DETAILED ACTION
This communication is in response to the Amendments and Arguments filed on 12/01/2025.
Claim(s) 1-11 are pending and have been examined. Hence, this action has been made FINAL.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments and Amendments
Amendments to the claims by the Applicant have been considered and addressed below.
With respect to the 35 USC § 101, 102, and 103 rejections, the Applicant provides several arguments in which the Examiner will respond accordingly, below.
35 USC § 101 rejection(s)
Arguments:
Claims 1-11 are rejected under 35 USC 101 as being directed to an abstract idea (mental process).
MPEP 2106.04(a)(2)III(A) defines what a mental process is for purposes of
determining whether an abstract idea is present. That MPEP section cites a number of Federal Circuit decisions finding that a mental process cannot be present if the claim limitations at issue cannot be performed entirely within the human mind. For example, that section provides in part:
Claims do not recite a mental process when they do not contain limitations that can practically be performed in the human mind, for instance when the human
mind is not equipped to perform the claim limitations. See SRI Int'l, Inc. v.
Cisco Systems, Inc., 930F.3d 1295, 1304 (Fed. Cir. 2019) (declining to identify the claimed collection and analysis of network data as abstract because "the human mind is not equipped to detect suspicious activity by using network monitors and analyzing network packets as recited by the claims")
It is Applicant's position that the tasks required in the present claims are too complicated to be performed entirely within the human mind and therefore the claims are not directed to a mental process.
Moreover, it is believed that the claimed invention makes the remarkable technical contribution corresponding to an inventive concept, which renders the amended claims, as a whole, as amounting to significantly more than the judicial exception, as affirmed by the holding in Contour IP Holding LLP v. GoPro, Inc., No. 2022-1654 (Fed. Cir. Sept. 9, 2024).
Examiner’s Response to Arguments:
Applicant' s arguments, with respect to the rejection(s) of independent claim(s) 1-11 under 35 USC 101 have been fully considered but are not persuasive.
The Examiner respectfully disagrees with the arguments of “the present claims [being] too complicated to be performed entirely within the human mind” and of the “claimed inventions mak[ing] the remarkable technical contribution corresponding to an inventive concept…” The Examiner notes that under the broadest reasonable interpretation, the limitations are considered to be practically performed by a human (mentally and/or using pen and paper) and refers the Applicant to the following MPEP sentions:
MPEP 2106.04(a)(2)(C)(III):
MENTAL PROCESSES
Nor do the courts distinguish between claims that recite mental processes performed by humans and claims that recite mental processes performed on a computer. As the Federal Circuit has explained, "[c]ourts have examined claims that required the use of a computer and still found that the underlying, patent-ineligible invention could be performed via pen and paper or in a person’s mind." Versata Dev. Group v. SAP Am., Inc., 793 F.3d 1306, 1335, 115 USPQ2d 1681, 1702 (Fed. Cir. 2015). See also Intellectual Ventures I LLC v. Symantec Corp., 838 F.3d 1307, 1318, 120 USPQ2d 1353, 1360 (Fed. Cir. 2016) (‘‘[W]ith the exception of generic computer-implemented steps, there is nothing in the claims themselves that foreclose them from being performed by a human, mentally or with pen and paper.’’); Mortgage Grader, Inc. v. First Choice Loan Servs. Inc., 811 F.3d 1314, 1324, 117 USPQ2d 1693, 1699 (Fed. Cir. 2016) (holding that computer-implemented method for "anonymous loan shopping" was an abstract idea because it could be "performed by humans without a computer"). Mental processes recited in claims that require computers are explained further below with respect to point C.
MPEP 2106.04(a)(2)(C)(III)(C):
A Claim That Requires a Computer May Still Recite a Mental Process
Claims can recite a mental process even if they are claimed as being performed on a computer. The Supreme Court recognized this in Benson, determining that a mathematical algorithm for converting binary coded decimal to pure binary within a computer’s shift register was an abstract idea. The Court concluded that the algorithm could be performed purely mentally even though the claimed procedures "can be carried out in existing computers long in use, no new machinery being necessary." 409 U.S at 67, 175 USPQ at 675. See also Mortgage Grader, 811 F.3d at 1324, 117 USPQ2d at 1699 (concluding that concept of "anonymous loan shopping" recited in a computer system claim is an abstract idea because it could be "performed by humans without a computer").
In evaluating whether a claim that requires a computer recites a mental process, examiners should carefully consider the broadest reasonable interpretation of the claim in light of the specification. For instance, examiners should review the specification to determine if the claimed invention is described as a concept that is performed in the human mind and applicant is merely claiming that concept performed 1) on a generic computer, or 2) in a computer environment, or 3) is merely using a computer as a tool to perform the concept. In these situations, the claim is considered to recite a mental process. )
Please see detailed analysis below (Prong Two) for more details on how the Examiner understands the independent claims do not recite additional elements that integrate the judicial exception into a practical application. Hence, not qualifying as patent eligible subject matter under 35 U.S.C. § 101.
Please refer to MPEP 2106.04(1): Eligibility Step 2A: Whether a Claim is Directed to a Judicial Exception: Prong One.
“Prong One asks does the claim recite an abstract idea, law of nature, or natural phenomenon? In Prong One examiners evaluate whether the claim recites a judicial exception, i.e. whether a law of nature, natural phenomenon, or abstract idea is set forth or described in the claim. While the terms "set forth" and "described" are thus both equated with "recite", their different language is intended to indicate that there are two ways in which an exception can be recited in a claim. For instance, the claims in Diehr, 450 U.S. at 178 n. 2, 179 n.5, 191-92, 209 USPQ at 4-5 (1981), clearly stated a mathematical equation in the repetitively calculating step, and the claims in Mayo, 566 U.S. 66, 75-77, 101 USPQ2d 1961, 1967-68 (2012), clearly stated laws of nature in the wherein clause, such that the claims "set forth" an identifiable judicial exception. Alternatively, the claims in Alice Corp., 573 U.S. at 218, 110 USPQ2d at 1982, described the concept of intermediated settlement without ever explicitly using the words "intermediated" or "settlement."”
“An example of a claim that recites a judicial exception is "A machine comprising elements that operate in accordance with F=ma." This claim sets forth the principle that force equals mass times acceleration (F=ma) and therefore recites a law of nature exception. Because F=ma represents a mathematical formula, the claim could alternatively be considered as reciting an abstract idea. Because this claim recites a judicial exception, it requires further analysis in Prong Two in order to answer the Step 2A inquiry. An example of a claim that merely involves, or is based on, an exception is a claim to "A teeter-totter comprising an elongated member pivotably attached to a base member, having seats and handles attached at opposing sides of the elongated member." This claim is based on the concept of a lever pivoting on a fulcrum, which involves the natural principles of mechanical advantage and the law of the lever. However, this claim does not recite these natural principles and therefore is not directed to a judicial exception (Step 2A: NO). Thus, the claim is eligible at Pathway B without further analysis.”
From this analysis, in Step 2A, Prong One, the Examiner has evaluated the independent claims accordingly and determined that the amended independent claims as drafted indeed describe a judicial exception (i.e., an abstract idea), which represent a mental process (which can be performed by a human with pen and paper).
More specifically, similar to what was discussed in the Non-Final Rejection mailed on 10/01/2025:
The independent claim(s) recite(s):
1. An information processing apparatus comprising:
at least one processor,
wherein the processor
acquires document data, and an item and a numerical value which are associated with each other, and
extracts, in a case in which a second numerical value within an allowable range including an acquired first numerical value is included in the document data, a phrase existing around the second numerical value included in the document data as a candidate for a synonymous term of the item.
10. An information processing method comprising:
via a processor provided in an information processing apparatus,
[the limitations as in claim 1, above]
11. A non-transitory computer-readable storage medium storing an information processing program for causing a processor provided in an information processing apparatus to execute a process comprising:
[the limitations as in claim 1, above]
This reads on a human (e.g., mentally and/or using pen and paper):
Receiving data (e.g., text) including an item (e.g., word) and a number associated to the item;
Identifying/selecting a phrase associated with a second number as a synonym to the item (e.g., word).
Please also refer to MPEP 2106.05(f)(2): Whether the claim invokes computers or other machinery merely as a tool to perform an existing process, and MPEP 2106.06(b): Clear Improvement to a Technology or to Computer Functionality.
Please refer to MPEP 2106.04(2): Eligibility Step 2A: Whether a Claim is Directed to a Judicial Exception: Prong Two.
“Prong Two asks does the claim recite additional elements that integrate the judicial exception into a practical application? In Prong Two, examiners evaluate whether the claim as a whole integrates the exception into a practical application of that exception. If the additional elements in the claim integrate the recited exception into a practical application of the exception, then the claim is not directed to the judicial exception (Step 2A: NO) and thus is eligible at Pathway B. This concludes the eligibility analysis. If, however, the additional elements do not integrate the exception into a practical application, then the claim is directed to the recited judicial exception (Step 2A: YES), and requires further analysis under Step 2B (where it may still be eligible if it amounts to an “inventive concept”). For more information on how to evaluate whether a judicial exception is integrated into a practical application, see MPEP § 2106.04(d)(2).”
From this analysis, in Step 2A, Prong Two, the Examiner has evaluated the independent claims accordingly and determined that the amended independent claims as drafted that the claims as a whole do not include additional elements that integrate the exception into a practical application of that exception. (i.e., an abstract idea). As discussed in the Non-Final Rejection mailed on 10/01/2025:
This judicial exception is not integrated into a practical application because for example: claim 1 recites “information processing apparatus” and “processor” while claim 10 also recites “information processing apparatus” and claim 11 recites a “non-transitory computer-readable storage medium”, “information processing program” and “processor”. As an example, in [0042] of the as filed specification, it is disclosed that “It should be noted that, in the embodiment described above, for example, as a hardware structure of a processing unit that executes various types of processing such as each functional unit of the information processing apparatus 10, various processors shown below can be used. As described above, in addition to the CPU that is a general-purpose processor that executes software (program) to function as various processing units, the various processors include a programmable logic device (PLD) that is a processor of which a circuit configuration can be changed after manufacture, such as a field programmable gate array (FPGA), and a dedicated electric circuit that is a processor having a circuit configuration that is designed for exclusive use in order to execute specific processing, such as an application specific integrated circuit (ASIC).” Therefore, a general-purpose computer or computing device is described and mainly used as an application thereof. Accordingly, these additional elements do not integrate the abstract idea into a practical idea because it does not impose any meaningful limits on practicing the abstract idea.
Please also refer to MPEP 2106.05(f)(2): Whether the claim invokes computers or other machinery merely as a tool to perform an existing process.
Finally, please refer to MPEP 2106.05(A): Relevant Considerations For Evaluating Whether Additional Elements Amount To An Inventive Concept
“Limitations that the courts have found not to be enough to qualify as "significantly more" when recited in a claim with a judicial exception include:
i. Adding the words "apply it" (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, e.g., a limitation indicating that a particular function such as creating and maintaining electronic records is performed by a computer, as discussed in Alice Corp., 573 U.S. at 225-26, 110 USPQ2d at 1984 (see MPEP § 2106.05(f));
ii. Simply appending well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception, e.g., a claim to an abstract idea requiring no more than a generic computer to perform generic computer functions that are well-understood, routine and conventional activities previously known to the industry, as discussed in Alice Corp., 573 U.S. at 225, 110 USPQ2d at 1984 (see MPEP § 2106.05(d));”
From this analysis, in Step 2B, the Examiner has evaluated the independent claims accordingly and determined that the independent claims as drafted have limitations that the courts have found not to be enough to qualify as "significantly more" when recited in a claim with a judicial exception. Similar to what was discussed in the Non-Final Rejection mailed on 10/01/2025:
The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements of using a computer is listed as a general computing device as noted. The claim is not patent eligible.
For more details, please refer to the 35 U.S.C. § 101 rejections for claims 1-11, below.
35 USC § 102 and 103 rejection(s)
Arguments:
Claims 1, 6, 7, 10 and 11 are rejected under 35 USC 102 as anticipated by Mikolov (USP 11,809,824).
Claims 2-5, 8 and 9 are rejected under 35 USC as obvious over Mikolov in view of Agatsuma (US 2024/0242026 A1).
As set forth below, it is believed that the originally filed claims are not anticipated by Mikolov or obvious over Mikolov in view of Agatsuma.
Mikolov (paragraphs [0003] and [0015]) relates to a system for converting one or more words within a sequence of words, such as sentences or phrases, into numerical representations in order to predict one or more unknown words in the sequence. Mikolov discloses receiving an input including an input word, generating a word score from the input word to obtain a numerical representation, and using the obtained numerical representations to reveal semantic and syntactic similarities or relationships between words.
However, Mikolov relates to a technique for expressing semantic similarity between words as numerical vectors, and does not disclose, as recited in independent claim 1, obtaining an actual "numerical value" from "document data," and, when "a second numerical value within an allowable range including an acquired first numerical value is included in the document data", extracting "a phrase existing around the second numerical value included in the document data as a candidate for a synonymous term of the item."
In addition, the distance in the vector space of Mikolov indicates semantic proximity, and is conceptually and technically different from numerical greater-than/less-than relationships or range comparisons, as in the present invention.
Further, while Mikolov "learns that words with similar surrounding
vocabularies tend to have similar meanings," the present invention is based on a different idea: when an item name and a number satisfy a certain condition, and that number appears in text, phrases related to the item name should appear around that number. Synonyms are then extracted from those surrounding phrases. Because of this, the basic design philosophy is fundamentally different.
Further, Mikolov deals with all words in a sentence, but the present invention uses the number in an "pair of item and number", as the key. Mikolov also does not disclose any technique for performing a range search based on that number, or extracting the surrounding phrases based on such numeric ranges.
Additionally, Mikolov effectively treats words that share similar surrounding words as similar words, while the present invention extracts phrases appearing around a number as synonym candidates and then evaluates synonymy among those phrases. Namely, Mikolov compares words to each other, whereas the present invention evaluates relationships among phrases that appear around numerical value and their relevance to the item name. The objects being compared are therefore completely different. It is noted that Mikolov's comparison of target words is implicit rather than explicit.
Mikolov processes each word in a sentence by taking that word as the center and then evaluating each pair consisting of the word and its surrounding words one by one. In contrast, in the present invention, even when the same numerical value appears multiple times in a document, all phrases appearing around every occurrence of that number are broadly collected and combined into a single set (list). The invention then extracts and evaluates representative phrases as a synonym list based on factors such as frequency and the number of occurrences within this set. In other words, unlike Mikolov's approach, which performs sequential processing on a "word-and-context pair" basis, the present invention aggregates all surrounding phrases on an "item-number basis" and evaluates patterns and characteristics across the entire aggregated set. This reveals a clear difference in both the processing unit and the underlying extraction concept.
In light of the above arguments, it is respectfully requested that the present rejections be withdrawn.
Examiner’s Response to Arguments:
Arguments have been considered but these are not persuasive. The Examiner respectfully disagrees with the arguments of Mikolov et al. not teaching “obtaining an actual "numerical value" from "document data," and, when "a second numerical value within an allowable range including an acquired first numerical value is included in the document data", extracting "a phrase existing around the second numerical value included in the document data as a candidate for a synonymous term of the item."”. The Examiner notes that the claim language as drafted reads differently and is broader than what is being argued (please see below for the actual language in the claims):
acquires document data, and an item and a numerical value which are associated with each other, (Here, the Examiner notes that there is no limitation with regards to the numeral value being obtained from the document data.)
extracts, in a case in which a second numerical value within an allowable range including an acquired first numerical value is included in the document data, a phrase existing around the second numerical value included in the document data as a candidate for a synonymous term of the item (Here, the Examiner notes that the acquired first value could be computed or obtained, and the numerical representation associated with the position read on the numerical value being included in the document data. The Examiner refers the Applicant to the citations below, incorporated for reference.)
¶ Col. 1, lines 33-51 and ¶ Col. 8, lines 63-67 citations as in limitation above and further ¶ starting on Col. 8, line 57: “(33) Once the word prediction system 100 and the word prediction system 300 have been trained and the parameters of the embedding function 106 and the embedding function 306 have been adjusted, the numeric representations produced by the embedding functions can be used for a variety of purposes other than as input to a classifier. For example, by training the word prediction system 100 or the word prediction system 300 to generate trained values of embedding function parameters for the embedding function 106 or the embedding function 306, e.g., as described below with reference to FIG. 5, the numeric representations produced by the embedding functions can encode many useful regularities. That is, the positions of the representations in the high-dimensional space can reflect syntactic similarities, e.g., showing that, by virtue of the positions of the numerical representations of each word in the space, words that are similar to the word “small” include the words “smaller,” “tiny,” smallest,” and so on, and semantic similarities, e.g., showing that the word “queen” is similar to the words “king” and “prince.” Furthermore, because of the encoded regularities, the numeric representations may show that the word “king” is similar to the word “queen” in the same sense as the word “prince” is similar to the word “princess,” and alternatively that the word “king” is similar to the word “prince” as the word “queen” is similar to the word “princess.” Advantageously, operations can be performed on the numeric representations to identify words that have a desired relationship to other words. In particular, vector subtraction and vector addition operations performed on floating point vectors generated by an embedding function in accordance with trained values of the parameters of the embedding function can be used to determine relationships between words. For example, in order to identify a word that has a similar relationship to a word A as a word B has to a word C, the following operation may be performed on the vectors representing words A, B, and C: vector(A)−vector(B)+vector(C). For example, the operation vector(“King”)−vector(“Man”)+vector(“Woman”) may result in a vector that is closest to the vector representation of the word “Queen.”
Here, the Examiner notes that the determination of the vector of the word “Queen” being closest to the vector representation of the word “King” based on their positions in the high-dimensional space and their semantic similarities would read on the extraction of a phrase/word as a synonymous term candidate. Additionally, the allowable range is considered to be a very broad limitation, hence, vectors of words within the same high-dimensional space would read on it.).
Similarly, the Examiner respectfully disagrees with the arguments (listed below) of Mikolov et al. not teaching limitations argued but not particularly claimed, considering that the language claim as drafted is much broader (as incorporated above for reference):
distance in the vector space of Mikolov indicates semantic proximity, and is conceptually and technically different from numerical greater-than/less-than relationships or range comparisons (The Examiner notes that claim language is broader than what is being argued. What is the condition? The Examiner notes, again, that the limitation of an “allowable range” is too broad. Hence, vectors of words within the same high-dimensional space would read on it.)
when an item name and a number satisfy a certain condition, and that number appears in text, phrases related to the item name should appear around that number. Synonyms are then extracted from those surrounding phrases (What is the condition? The Examiner notes, again, that the limitation of an “allowable range” is too broad. Hence, vectors of words within the same high-dimensional space would read on it.)
technique for performing a range search based on that number, or extracting the surrounding phrases based on such numeric ranges (The Examiner notes that claim language is broader than what is being argued. Also, what is the technique for performing search based on the numerical value and potentially defining the allowable range?)
extracts phrases appearing around a number as synonym candidates and then evaluates synonymy among those phrases (The Examiner notes that claim language is broader than what is being argued. Also, Applicant is further referred to Mikolov et al. ¶ at Col. 5, line 41).
evaluates relationships among phrases that appear around numerical value and their relevance to the item name
even when the same numerical value appears multiple times in a document, all phrases appearing around every occurrence of that number are broadly collected and combined into a single set (list) and then extracts and evaluates representative phrases as a synonym list based on factors such as frequency and the number of occurrences within this set (The Examiner notes that claim language is broader than what is being argued in the independent claims.)
aggregates all surrounding phrases on an "item-number basis" and evaluates patterns and characteristics across the entire aggregated set (The Examiner notes that claim language is broader than what is being argued in the independent claims.)
For more details, please refer to updated 35 U.S.C. § 103 rejections for claims 1-11, below.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim(s) 1-11 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. More specifically directed to the abstract idea grouping of: mental process.
The independent claim(s) recite(s):
1. An information processing apparatus comprising:
at least one processor,
wherein the processor
acquires document data, and an item and a numerical value which are associated with each other, and
extracts, in a case in which a second numerical value within an allowable range including an acquired first numerical value is included in the document data, a phrase existing around the second numerical value included in the document data as a candidate for a synonymous term of the item.
10. An information processing method comprising:
via a processor provided in an information processing apparatus,
[the limitations as in claim 1, above]
11. A non-transitory computer-readable storage medium storing an information processing program for causing a processor provided in an information processing apparatus to execute a process comprising:
[the limitations as in claim 1, above]
This reads on a human (e.g., mentally and/or using pen and paper):
Receiving data (e.g., text) including an item (e.g., word) and a number associated to the item;
Identifying/selecting a phrase associated with a second number as a synonym to the item (e.g., word).
This judicial exception is not integrated into a practical application because for example: claim 1 recites “information processing apparatus” and “processor” while claim 10 also recites “information processing apparatus” and claim 11 recites a “non-transitory computer-readable storage medium”, “information processing program” and “processor”. As an example, in [0042] of the as filed specification, it is disclosed that “It should be noted that, in the embodiment described above, for example, as a hardware structure of a processing unit that executes various types of processing such as each functional unit of the information processing apparatus 10, various processors shown below can be used. As described above, in addition to the CPU that is a general-purpose processor that executes software (program) to function as various processing units, the various processors include a programmable logic device (PLD) that is a processor of which a circuit configuration can be changed after manufacture, such as a field programmable gate array (FPGA), and a dedicated electric circuit that is a processor having a circuit configuration that is designed for exclusive use in order to execute specific processing, such as an application specific integrated circuit (ASIC).” Therefore, a general-purpose computer or computing device is described and mainly used as an application thereof. Accordingly, these additional elements do not integrate the abstract idea into a practical idea because it does not impose any meaningful limits on practicing the abstract idea.
The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional elements of using a computer is listed as a general computing device as noted. The claim is not patent eligible.
With respect to claim 2, the claim(s) recite:
2. The information processing apparatus according to claim 1,
wherein the processor
generates a synonymous term list from the candidates for the synonymous term based on a statistical value of the extracted phrase in accumulation data in which a plurality of sets of the items and the numerical values are accumulated.
This reads on a human (e.g., mentally and/or using pen and paper):
Writing down a list of synonyms for words present in the data received based on statistics (e.g., mathematical steps)
No additional limitations are present.
With respect to claim 3, the claim(s) recite:
3. The information processing apparatus according to claim 2,
wherein the statistical value is the number of times of appearance of the extracted phrase in the accumulation data, and
the processor adds, to the synonymous term list, a phrase of which the statistical value is equal to or larger than a threshold value among the extracted phrases.
This reads on a human (e.g., mentally and/or using pen and paper):
Wherein the based on statistics consists of number of times a word appears in the data (i.e., text)
Further writes down terms on the list.
No additional limitations are present.
With respect to claim 4, the claim(s) recite:
4. The information processing apparatus according to claim 2,
wherein the statistical value is the number of times of appearance of the extracted phrase in the accumulation data, and
the processor adds, to the synonymous term list, a phrase of which the statistical value is relatively large among candidates for the extracted phrase.
This reads on a human (e.g., mentally and/or using pen and paper):
Wherein the based on statistics consists of number of times a word appears in the data (i.e., text) (i.e., respective steps)
Further writes down terms on the list.
No additional limitations are present.
With respect to claim 5, the claim(s) recite:
5. The information processing apparatus according to claim 2,
wherein the processor
acquires a reference value corresponding to the item, and
extracts, in a case in which a word corresponding to a magnitude relationship of the acquired numerical value with respect to the reference value is included in the document data, a phrase existing around the word included in the document data, and
the statistical value includes a statistical value of a phrase existing around the word in the accumulation data.
This reads on a human (e.g., mentally and/or using pen and paper):
Receiving a reference value (e.g., predefined/known range)
Determining or identifying a word that meets said value
Using statistical values (i.e., mathematical steps) to compare with surrounding words/phrases.
No additional limitations are present.
With respect to claim 6, the claim(s) recite:
6. The information processing apparatus according to claim 1,
wherein the processor
performs weighting by setting a weight coefficient of a statistical value of a phrase extracted by applying a unit in the numerical value and including the second numerical value and the unit in the document data to a value larger than a weight coefficient of a statistical value of a phrase extracted by including only the second numerical value in the document data.
This reads on a human (e.g., mentally and/or using pen and paper):
Assigning weights or numerical values to phrases/words and based on thresholds selecting a particular phrase/words (i.e., predefined set of rules/steps).
No additional limitations are present.
With respect to claim 7, the claim(s) recite:
7. The information processing apparatus according to claim 1,
wherein the processor
extracts, in a case in which the phrase existing around the second numerical value is extracted, a plurality of phrases having different lengths or positions as the candidates for the synonymous term.
This reads on a human (e.g., mentally and/or using pen and paper):
Identifying a phrases with different lengths/locations in data (e.g., text) as potential synonyms.
No additional limitations are present.
With respect to claim 8, the claim(s) recite:
8. The information processing apparatus according to claim 2,
wherein the processor
performs weighting of the statistical value of the phrase based on a statistical value of the acquired numerical value in the accumulation data.
This reads on a human (e.g., mentally and/or using pen and paper):
Assigning weights or numerical values to phrases/words and based on statistics (i.e., mathematical steps and/or predefined set of rules/steps).
No additional limitations are present.
With respect to claim 9, the claim(s) recite:
9. The information processing apparatus according to claim 2,
wherein the processor
derives a degree of similarity between the acquired item and the extracted phrase, and
performs weighting by setting a weight coefficient of the statistical value of the phrase of which the degree of similarity is equal to or larger than a certain value to a value larger than a weight coefficient of the statistical value of the phrase of which the degree of similarity is smaller than the certain value.
This reads on a human (e.g., mentally and/or using pen and paper):
Determining a similarity degree/value (predefined set of rules/steps)
Assigning weights or numerical values to phrases/words and based on thresholds (i.e., predefined set of rules/steps).
No additional limitations are present.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 1, 6-7, and 10-11 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Mikolov et al. (US 11809824 B1).
As to independent claims 1 and 10-11, Mikolov et al. teaches:
1. An information processing apparatus (see ¶ Col. 1, lines 33-51: “(3) In general, one innovative aspect of the subject matter described in this specification can be embodied in a system that includes a classifier implemented in one or more computers, comprising […] and instructions to process each word in a vocabulary of words using the embedding function layer to obtain a respective numeric representation of each word in the vocabulary in the high-dimensional space and to associate each word in the vocabulary with the respective numeric representation of the word in the high-dimensional space.”
and ¶ Col. 2, lines 51-54: “(8) Other embodiments of this aspect include corresponding systems, apparatus, and computer programs, configured to perform the actions of the methods, encoded on computer storage devices.”) comprising:
at least one processor (see ¶ Col. 11, lines 23-50: “(45) Computers suitable for the execution of a computer program include, by way of example, can be based on general or special purpose microprocessors or both, or any other kind of central processing unit. Generally, a central processing unit will receive instructions and data from a read only memory or a random access memory or both. The essential elements of a computer are a central processing unit for performing or executing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto optical disks, or optical disks. However, a computer need not have such devices. Moreover, a computer can be embedded in another device, e.g., a mobile telephone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a Global Positioning System (GPS) receiver, or a portable storage device, e.g., a universal serial bus (USB) flash drive, to name just a few. (46) Computer readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media and memory devices, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices; magnetic disks, e.g., internal hard disks or removable disks; magneto optical disks; and CD ROM and DVD-ROM disks. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.”), wherein the processor
acquires document data, and an item and a numerical value which are associated with each other (see ¶ Col. 1, lines 33-51: “(3) In general, one innovative aspect of the subject matter described in this specification can be embodied in a system that includes a classifier implemented in one or more computers, comprising: an embedding function layer configured to receive an input comprising a plurality of words that surround an unknown word in a sequence of words and map the plurality of words into a numeric representation in a high-dimensional space;…
(47)… In addition, a computer can interact with a user by sending documents to and receiving documents from a device that is used by the user; for example, by sending web pages to a web browser on a user's client device in response to requests received from the web browser.
and ¶ Col. 8, lines 63-67: “… In addition, a computer can interact with a user by sending documents to and receiving documents from a device that is used by the user; for example, by sending web pages to a web browser on a user's client device in response to requests received from the web browser.”), and
extracts, in a case in which a second numerical value within an allowable range including an acquired first numerical value is included in the document data, a phrase existing around the second numerical value included in the document data as a candidate for a synonymous term of the item (see ¶ Col. 1, lines 33-51 and ¶ Col. 8, lines 63-67 citations as in limitation above and further ¶ starting on Col. 8, line 57: “(33) Once the word prediction system 100 and the word prediction system 300 have been trained and the parameters of the embedding function 106 and the embedding function 306 have been adjusted, the numeric representations produced by the embedding functions can be used for a variety of purposes other than as input to a classifier. For example, by training the word prediction system 100 or the word prediction system 300 to generate trained values of embedding function parameters for the embedding function 106 or the embedding function 306, e.g., as described below with reference to FIG. 5, the numeric representations produced by the embedding functions can encode many useful regularities. That is, the positions of the representations in the high-dimensional space can reflect syntactic similarities, e.g., showing that, by virtue of the positions of the numerical representations of each word in the space, words that are similar to the word “small” include the words “smaller,” “tiny,” smallest,” and so on, and semantic similarities, e.g., showing that the word “queen” is similar to the words “king” and “prince.” Furthermore, because of the encoded regularities, the numeric representations may show that the word “king” is similar to the word “queen” in the same sense as the word “prince” is similar to the word “princess,” and alternatively that the word “king” is similar to the word “prince” as the word “queen” is similar to the word “princess.” Advantageously, operations can be performed on the numeric representations to identify words that have a desired relationship to other words. In particular, vector subtraction and vector addition operations performed on floating point vectors generated by an embedding function in accordance with trained values of the parameters of the embedding function can be used to determine relationships between words. For example, in order to identify a word that has a similar relationship to a word A as a word B has to a word C, the following operation may be performed on the vectors representing words A, B, and C: vector(A)−vector(B)+vector(C). For example, the operation vector(“King”)−vector(“Man”)+vector(“Woman”) may result in a vector that is closest to the vector representation of the word “Queen.”
Here, the Examiner notes that the determination of the vector of the word “Queen” being closest to the vector representation of the word “King” based on their positions in the high-dimensional space and their semantic similarities would read on the extraction of a phrase/word as a synonymous term candidate. Additionally, the allowable range is considered to be a very broad limitation, hence, vectors of words within the same high-dimensional space would read on it.).
10. An information processing method (see ¶ Col. 1, lines 33-51: “(3) In general, one innovative aspect of the subject matter described in this specification can be embodied in a system that includes a classifier implemented in one or more computers, comprising […] and instructions to process each word in a vocabulary of words using the embedding function layer to obtain a respective numeric representation of each word in the vocabulary in the high-dimensional space and to associate each word in the vocabulary with the respective numeric representation of the word in the high-dimensional space.”
and ¶ Col. 2, lines 51-54: “(8) Other embodiments of this aspect include corresponding systems, apparatus, and computer programs, configured to perform the actions of the methods, encoded on computer storage devices.”) comprising:
via a processor provided in an information processing apparatus (see ¶ Col. 11, lines 23-50 citation as in claim 1, above.),
[the limitations as in claim 1, above.]
11. A non-transitory computer-readable storage medium storing an information processing program (see ¶ Col. 10, lines 35-40: “…Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible non transitory program carrier for execution by, or to control the operation of, data processing apparatus…”) for causing a processor provided in an information processing apparatus to execute a process (see ¶ Col. 11, lines 23-50 citation as in claim 1, above.) comprising:
[the limitations as in claim 1, above.]
Regarding claim 6, Mikolov et al. teaches the limitations as in claim 1, above.
Mikolov et al. further teaches:
6. The information processing apparatus according to claim 1,
wherein the processor
performs weighting by setting a weight coefficient of a statistical value of a phrase extracted by applying a unit in the numerical value and including the second numerical value and the unit in the document data to a value larger than a weight coefficient of a statistical value of a phrase extracted by including only the second numerical value in the document data (see ¶ Col. 1, lines 33-51 and ¶ Col. 8, lines 63-67 citations as in limitation above and further ¶ starting at Col. 2, line 51: “…processing each word in the vocabulary using the embedding function in accordance with the trained values of the embedding function parameters to generate a respective numerical representation of each word in the vocabulary; and associating each word in the vocabulary with the respective numeric representation of the word in the high-dimensional space.”,
¶ starting on Co. 7, line 19: “(27) Each of the classifiers 310 receives the numeric representation generated by the embedding function 306 and predicts a value for each field of a respective word score vector in accordance with values of a respective set of classifier parameters. Generally, each of the classifiers 310 will have different values of the classifier parameters. Each word score vector corresponds to a respective position in the sequence of words. For example, the word score vector 312 includes values for each of the predetermined set of words that is a prediction of how likely it is that the corresponding word is the word at position t-N in the sequence. The classifiers 310 can be any multiclass or multilabel classifier, e.g., a multiclass logistic regression classifier, a multiclass support vector machine classifier, a Bayesian classifier, and so on. In some implementations, instead of the classifiers 310, the concept term scoring system 300 can include ranking functions that each order the words based on the numeric representation generated by the embedding function 106, i.e., in order of predicted likelihood of being the word at the corresponding position. The ranking function may be, e.g., a hinge-loss ranking function, a pairwise ranking function, and so on.”, and
¶ starting on Col. 8, line 57: “(33) Once the word prediction system 100 and the word prediction system 300 have been trained and the parameters of the embedding function 106 and the embedding function 306 have been adjusted, the numeric representations produced by the embedding functions can be used for a variety of purposes other than as input to a classifier. For example, by training the word prediction system 100 or the word prediction system 300 to generate trained values of embedding function parameters for the embedding function 106 or the embedding function 306, e.g., as described below with reference to FIG. 5, the numeric representations produced by the embedding functions can encode many useful regularities. That is, the positions of the representations in the high-dimensional space can reflect syntactic similarities, e.g., showing that, by virtue of the positions of the numerical representations of each word in the space, words that are similar to the word “small” include the words “smaller,” “tiny,” smallest,” and so on, and semantic similarities, e.g., showing that the word “queen” is similar to the words “king” and “prince.” Furthermore, because of the encoded regularities, the numeric representations may show that the word “king” is similar to the word “queen” in the same sense as the word “prince” is similar to the word “princess,” and alternatively that the word “king” is similar to the word “prince” as the word “queen” is similar to the word “princess.” Advantageously, operations can be performed on the numeric representations to identify words that have a desired relationship to other words. In particular, vector subtraction and vector addition operations performed on floating point vectors generated by an embedding function in accordance with trained values of the parameters of the embedding function can be used to determine relationships between words. For example, in order to identify a word that has a similar relationship to a word A as a word B has to a word C, the following operation may be performed on the vectors representing words A, B, and C: vector(A)−vector(B)+vector(C). For example, the operation vector(“King”)−vector(“Man”)+vector(“Woman”) may result in a vector that is closest to the vector representation of the word “Queen.”).
Regarding claim 7, Mikolov et al. teaches the limitations as in claim 1, above.
Mikolov et al. further teaches:
7. The information processing apparatus according to claim 1,
wherein the processor
extracts, in a case in which the phrase existing around the second numerical value is extracted, a plurality of phrases having different lengths or positions as the candidates for the synonymous term (see ¶ starting on Co. 7, line 19: “(27) Each of the classifiers 310 receives the numeric representation generated by the embedding function 306 and predicts a value for each field of a respective word score vector in accordance with values of a respective set of classifier parameters. Generally, each of the classifiers 310 will have different values of the classifier parameters. Each word score vector corresponds to a respective position in the sequence of words. For example, the word score vector 312 includes values for each of the predetermined set of words that is a prediction of how likely it is that the corresponding word is the word at position t-N in the sequence. The classifiers 310 can be any multiclass or multilabel classifier, e.g., a multiclass logistic regression classifier, a multiclass support vector machine classifier, a Bayesian classifier, and so on. In some implementations, instead of the classifiers 310, the concept term scoring system 300 can include ranking functions that each order the words based on the numeric representation generated by the embedding function 106, i.e., in order of predicted likelihood of being the word at the corresponding position. The ranking function may be, e.g., a hinge-loss ranking function, a pairwise ranking function, and so on.”, ¶ starting at Col. 5, line 41: “(18) …As a simplified example, for the ordered list {"Atlanta", "Hotel"}, the parallel embedding function may map "Atlanta" to a vector [0.1, 0.2, 0.3] and "Hotel" to [0.4, 0.5, 0.6], and then output the sum of the two vectors, i.e., [0.5, 0.7, 0.9].”
¶ starting at Col. 7, line 56: “(31) The system processes the word using an embedding function (step 404) to generate a numeric representation of the word in a high-dimensional space. The embedding function maps the word to a continuous high-dimensional representation, e.g., to a high-dimensional vector of floating point numbers. For example, the embedding function may map the word `cat` to a vector [0.1, 0.5, 0.2] and the word `tablet` to a vector [0.3, 0.9, 0.0], based on current parameter values of the embedding function, e.g., as stored in a lookup table.”).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 2-5 and 8-9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Mikolov et al. (US 11809824 B1) as applied to claims 1 above, and further in view of Agatsuma et al. (US 20240242026 A1).
Regarding claim 2, Mikolov et al. teaches the limitations as in claim 1, above.
However, Mikolov et al. does not explicitly teach, but Agatsuma et al. does teach:
2. The information processing apparatus according to claim 1,
wherein the processor
generates a synonymous term list from the candidates for the synonymous term based on a statistical value of the extracted phrase in accumulation data in which a plurality of sets of the items and the numerical values are accumulated (see ¶ [0046]: “The storage unit 110 stores, as main information (data), a document information table 111, a word category list 112, a word category determination model 113, a word table 114, a synonym candidate table 115, a mismatch substring table 116, a threshold table 117, a substring correct/incorrect table 118, a synonym dictionary 121, and a non-synonym dictionary 122. Details thereof will be described later.”, ¶ [0064]: “The synonym candidate generation unit 140 obtains a feature (hereinafter, referred to as a “relationship feature”) indicating relationship between two words forming a word pair (hereinafter, also referred to as a “synonym candidate”) that is a combination of two words having the same category and managed in the word table 114 for the word pair, and stores the word pair and the relationship feature of the word pair in association with each other in the synonym candidate table 115. The synonym candidate generation unit 140 uses, as the relationship feature, for example, a co-occurrence frequency of the word pair acquired by applying a machine learning model (word2vec or the like) from the document data of the document information table 111, an editing distance of the word pair, a category association probability of the word pair, the number of appearances of the word pair, and a sentence (text data) of an extraction source of each word of the word pair. Note that the relationship feature is not necessarily limited thereto. The content of the synonym candidate table 115 may be set by a user via a user interface provided by the user apparatus 2.”
¶ [0077]: “FIG. 7 illustrates an example of the synonym candidate table 115 managed by the storage unit 110 illustrated in FIG. 1. Information regarding a word pair that is synonym candidates is managed in the synonym candidate table 115. The word pair is a combination of two words belonging to the same category extracted from words in the word table 114. As illustrated in FIG. 7, the exemplified synonym candidate table 115 includes a plurality of records each having respective items such as a word A 1151, a word B 1152, correct/incorrect information 1153, a word category 1154, a co-occurrence frequency 1155, an editing distance 1156, a category association probability 1157, the number of appearances 1158, and an extraction source text 1159.”
and ¶ [0080]: “Specific values of relationship features (a co-occurrence frequency, an editing distance, a category association probability, the number of appearances, extraction source text) are stored in the co-occurrence frequency 1155, the editing distance 1156, the category association probability 1157, the number of appearances 1158, and the extraction source text 1159. A co-occurrence frequency of the word A and the word B calculated by using a machine learning model or the like is stored in the co-occurrence frequency 1155...”).
Mikolov et al. and Agatsuma et al. are considered to be analogous to the claimed invention because they are in the same field of endeavor in document/data processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Mikolov et al. to incorporate the teachings of Agatsuma et al. of wherein the processor generates a synonymous term list from the candidates for the synonymous term based on a statistical value of the extracted phrase in accumulation data in which a plurality of sets of the items and the numerical values are accumulated which provides the benefit of efficiently extracting synonyms from document data with high accuracy (abstract of Agatsuma et al.).
Regarding claim 3, Mikolov et al. in combination with Agatsuma et al. teaches the limitations as in claim 2, above.
Agatsuma et al. further teaches:
3. The information processing apparatus according to claim 2,
wherein the statistical value is the number of times of appearance of the extracted phrase in the accumulation data (see ¶ [0064, 0077, and 0080] citations as in claim 2, above. More specifically: “¶ [0077]: “FIG. 7 illustrates an example of the synonym candidate table 115 managed by the storage unit 110 illustrated in FIG. 1. Information regarding a word pair that is synonym candidates is managed in the synonym candidate table 115. The word pair is a combination of two words belonging to the same category extracted from words in the word table 114. As illustrated in FIG. 7, the exemplified synonym candidate table 115 includes a plurality of records each having respective items such as a word A 1151, a word B 1152, correct/incorrect information 1153, a word category 1154, a co-occurrence frequency 1155, an editing distance 1156, a category association probability 1157, the number of appearances 1158, and an extraction source text 1159.”), and
the processor
adds, to the synonymous term list, a phrase of which the statistical value is equal to or larger than a threshold value among the extracted phrases (see ¶ [0062]: “…The synonym determination system 1 uses correct/incorrect information received from the user and synonym extraction rules (the threshold table 117 and the substring correct/incorrect table 118) generated by the synonym extraction rule generation unit 180, as a criterion for determining whether two words forming a word pair are synonyms or non-synonyms.”
and ¶ [0123]: “Subsequently, the synonym extraction rule applying unit 150 compares a relationship feature in the selected record with a threshold in the threshold table 117, and determines whether there is a relationship feature less than the threshold (S1614). Specifically, the synonym extraction rule applying unit 150 determines whether there is a relationship feature less than a corresponding threshold in the threshold table 117 among the relationship features (the co-occurrence frequency 1155, the editing distance 1156, the category association probability of each of the word A and the word B in the category association probability 1157, and the number of appearances of each of the word A and the word B in the number of appearances 1158) in the record. In the above determination, the synonym extraction rule applying unit 150 uses a value stored in the category association probability threshold 1172 in the threshold table 117 of the common category to which the word A and the word B belong for the thresholds of the category association probabilities of the word A and the word B. In a case where there is even one relationship feature less than the threshold (S1614: YES), the synonym extraction rule applying unit 150 registers the word pair of the selected record in the non-synonym dictionary 122 (S1621), and deletes the selected record from the synonym candidate table 115 (S1622). Thereafter, the process proceeds to S1620. On the other hand, in a case where there is no relationship feature less than the threshold (S1621: NO), the process proceeds to S1615.”).
Mikolov et al. and Agatsuma et al. are considered to be analogous to the claimed invention because they are in the same field of endeavor in document/data processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Mikolov et al. to incorporate the teachings of Agatsuma et al. of wherein the statistical value is the number of times of appearance of the extracted phrase in the accumulation data, and the processor adds, to the synonymous term list, a phrase of which the statistical value is equal to or larger than a threshold value among the extracted phrases which provides the benefit of efficiently extracting synonyms from document data with high accuracy (abstract of Agatsuma et al.).
Regarding claim 4, Mikolov et al. in combination with Agatsuma et al. teaches the limitations as in claim 2, above.
Agatsuma et al. further teaches:
4. The information processing apparatus according to claim 2,
wherein the statistical value is the number of times of appearance of the extracted phrase in the accumulation data (see ¶ [0064, 0077, and 0080] citations as in claim 2, above. More specifically: ¶ [0064]: “…. The synonym candidate generation unit 140 uses, as the relationship feature, for example, a co-occurrence frequency of the word pair acquired by applying a machine learning model (word2vec or the like) from the document data of the document information table 111, an editing distance of the word pair, a category association probability of the word pair, the number of appearances of the word pair, and a sentence (text data) of an extraction source of each word of the word pair. Note that the relationship feature is not necessarily limited thereto. The content of the synonym candidate table 115 may be set by a user via a user interface provided by the user apparatus 2.”
¶ [0077]: “FIG. 7 illustrates an example of the synonym candidate table 115 managed by the storage unit 110 illustrated in FIG. 1. Information regarding a word pair that is synonym candidates is managed in the synonym candidate table 115. The word pair is a combination of two words belonging to the same category extracted from words in the word table 114. As illustrated in FIG. 7, the exemplified synonym candidate table 115 includes a plurality of records each having respective items such as a word A 1151, a word B 1152, correct/incorrect information 1153, a word category 1154, a co-occurrence frequency 1155, an editing distance 1156, a category association probability 1157, the number of appearances 1158, and an extraction source text 1159.”
and ¶ [0080]: “Specific values of relationship features (a co-occurrence frequency, an editing distance, a category association probability, the number of appearances, extraction source text) are stored in the co-occurrence frequency 1155, the editing distance 1156, the category association probability 1157, the number of appearances 1158, and the extraction source text 1159. A co-occurrence frequency of the word A and the word B calculated by using a machine learning model or the like is stored in the co-occurrence frequency 1155...”)., and
the processor
adds, to the synonymous term list, a phrase of which the statistical value is relatively large among candidates for the extracted phrase (see ¶ [0062]: “…The synonym determination system 1 uses correct/incorrect information received from the user and synonym extraction rules (the threshold table 117 and the substring correct/incorrect table 118) generated by the synonym extraction rule generation unit 180, as a criterion for determining whether two words forming a word pair are synonyms or non-synonyms.”
and ¶ [0123]: “Subsequently, the synonym extraction rule applying unit 150 compares a relationship feature in the selected record with a threshold in the threshold table 117, and determines whether there is a relationship feature less than the threshold (S1614). Specifically, the synonym extraction rule applying unit 150 determines whether there is a relationship feature less than a corresponding threshold in the threshold table 117 among the relationship features (the co-occurrence frequency 1155, the editing distance 1156, the category association probability of each of the word A and the word B in the category association probability 1157, and the number of appearances of each of the word A and the word B in the number of appearances 1158) in the record. In the above determination, the synonym extraction rule applying unit 150 uses a value stored in the category association probability threshold 1172 in the threshold table 117 of the common category to which the word A and the word B belong for the thresholds of the category association probabilities of the word A and the word B. In a case where there is even one relationship feature less than the threshold (S1614: YES), the synonym extraction rule applying unit 150 registers the word pair of the selected record in the non-synonym dictionary 122 (S1621), and deletes the selected record from the synonym candidate table 115 (S1622). Thereafter, the process proceeds to S1620. On the other hand, in a case where there is no relationship feature less than the threshold (S1621: NO), the process proceeds to S1615.”).
Mikolov et al. and Agatsuma et al. are considered to be analogous to the claimed invention because they are in the same field of endeavor in document/data processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Mikolov et al. to incorporate the teachings of Agatsuma et al. of wherein the statistical value is the number of times of appearance of the extracted phrase in the accumulation data, and the processor adds, to the synonymous term list, a phrase of which the statistical value is relatively large among candidates for the extracted phrase which provides the benefit of efficiently extracting synonyms from document data with high accuracy (abstract of Agatsuma et al.).
Regarding claim 5, Mikolov et al. teaches the limitations as in claim 1, above.
Mikolov et al. further teaches:
5. The information processing apparatus according to claim 2,
wherein the processor
acquires a reference value corresponding to the item (see ¶ Col. 1, lines 33-51: “(3) In general, one innovative aspect of the subject matter described in this specification can be embodied in a system that includes a classifier implemented in one or more computers, comprising: an embedding function layer configured to receive an input comprising a plurality of words that surround an unknown word in a sequence of words and map the plurality of words into a numeric representation in a high-dimensional space;…
(47)… In addition, a computer can interact with a user by sending documents to and receiving documents from a device that is used by the user; for example, by sending web pages to a web browser on a user's client device in response to requests received from the web browser.
and ¶ Col. 8, lines 63-67: “… In addition, a computer can interact with a user by sending documents to and receiving documents from a device that is used by the user; for example, by sending web pages to a web browser on a user's client device in response to requests received from the web browser.”), and
extracts, in a case in which a word corresponding to a magnitude relationship of the acquired numerical value with respect to the reference value is included in the document data, a phrase existing around the word included in the document data (see ¶ Col. 1, lines 33-51 and ¶ Col. 8, lines 63-67 citations as in limitation above and further ¶ starting on Col. 8, line 57: “(33) Once the word prediction system 100 and the word prediction system 300 have been trained and the parameters of the embedding function 106 and the embedding function 306 have been adjusted, the numeric representations produced by the embedding functions can be used for a variety of purposes other than as input to a classifier. For example, by training the word prediction system 100 or the word prediction system 300 to generate trained values of embedding function parameters for the embedding function 106 or the embedding function 306, e.g., as described below with reference to FIG. 5, the numeric representations produced by the embedding functions can encode many useful regularities. That is, the positions of the representations in the high-dimensional space can reflect syntactic similarities, e.g., showing that, by virtue of the positions of the numerical representations of each word in the space, words that are similar to the word “small” include the words “smaller,” “tiny,” smallest,” and so on, and semantic similarities, e.g., showing that the word “queen” is similar to the words “king” and “prince.” Furthermore, because of the encoded regularities, the numeric representations may show that the word “king” is similar to the word “queen” in the same sense as the word “prince” is similar to the word “princess,” and alternatively that the word “king” is similar to the word “prince” as the word “queen” is similar to the word “princess.” Advantageously, operations can be performed on the numeric representations to identify words that have a desired relationship to other words. In particular, vector subtraction and vector addition operations performed on floating point vectors generated by an embedding function in accordance with trained values of the parameters of the embedding function can be used to determine relationships between words. For example, in order to identify a word that has a similar relationship to a word A as a word B has to a word C, the following operation may be performed on the vectors representing words A, B, and C: vector(A)−vector(B)+vector(C). For example, the operation vector(“King”)−vector(“Man”)+vector(“Woman”) may result in a vector that is closest to the vector representation of the word “Queen.”
Here, the Examiner notes that the determination of the vector of the word “Queen” being closest to the vector representation of the word “King” based on their positions in the high-dimensional space and their semantic similarities would read on the extraction of a phrase/word as a synonymous term candidate. Additionally, the allowable range is considered to be a very broad limitation, hence, vectors of words within the same high-dimensional space would read on it.),
However, Mikolov et al. does not explicitly teach, but Agatsuma et al. does teach:
the statistical value includes a statistical value of a phrase existing around the word in the accumulation data (see ¶ [0046]: “The storage unit 110 stores, as main information (data), a document information table 111, a word category list 112, a word category determination model 113, a word table 114, a synonym candidate table 115, a mismatch substring table 116, a threshold table 117, a substring correct/incorrect table 118, a synonym dictionary 121, and a non-synonym dictionary 122. Details thereof will be described later.”, ¶ [0064]: “The synonym candidate generation unit 140 obtains a feature (hereinafter, referred to as a “relationship feature”) indicating relationship between two words forming a word pair (hereinafter, also referred to as a “synonym candidate”) that is a combination of two words having the same category and managed in the word table 114 for the word pair, and stores the word pair and the relationship feature of the word pair in association with each other in the synonym candidate table 115. The synonym candidate generation unit 140 uses, as the relationship feature, for example, a co-occurrence frequency of the word pair acquired by applying a machine learning model (word2vec or the like) from the document data of the document information table 111, an editing distance of the word pair, a category association probability of the word pair, the number of appearances of the word pair, and a sentence (text data) of an extraction source of each word of the word pair. Note that the relationship feature is not necessarily limited thereto. The content of the synonym candidate table 115 may be set by a user via a user interface provided by the user apparatus 2.”
¶ [0077]: “FIG. 7 illustrates an example of the synonym candidate table 115 managed by the storage unit 110 illustrated in FIG. 1. Information regarding a word pair that is synonym candidates is managed in the synonym candidate table 115. The word pair is a combination of two words belonging to the same category extracted from words in the word table 114. As illustrated in FIG. 7, the exemplified synonym candidate table 115 includes a plurality of records each having respective items such as a word A 1151, a word B 1152, correct/incorrect information 1153, a word category 1154, a co-occurrence frequency 1155, an editing distance 1156, a category association probability 1157, the number of appearances 1158, and an extraction source text 1159.”
and ¶ [0080]: “Specific values of relationship features (a co-occurrence frequency, an editing distance, a category association probability, the number of appearances, extraction source text) are stored in the co-occurrence frequency 1155, the editing distance 1156, the category association probability 1157, the number of appearances 1158, and the extraction source text 1159. A co-occurrence frequency of the word A and the word B calculated by using a machine learning model or the like is stored in the co-occurrence frequency 1155...”).
Mikolov et al. and Agatsuma et al. are considered to be analogous to the claimed invention because they are in the same field of endeavor in document/data processing. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Mikolov et al. to incorporate the teachings of Agatsuma et al. of the statistical value includes a statistical value of a phrase existing around the word in the accumulation data which provides the benefit of efficiently extracting synonyms from document data with high accuracy (abstract of Agatsuma et al.).
Regarding claim 8, Mikolov et al. in combination with Agatsuma et al. teaches the limitations as in claim 2, above.
Mikolov et al. further teaches:
8. The information processing apparatus according to claim 2,
wherein the processor
performs weighting of the statistical value of the phrase based on a statistical value of the acquired numerical value in the accumulation data (¶ starting at Col. 2, line 51, ¶ starting on Co. 7, line 19, and ¶ starting on Col. 8, line 57 citations as in claim 6, above. More specifically: “¶ starting at Col. 2, line 51: “…processing each word in the vocabulary using the embedding function in accordance with the trained values of the embedding function parameters to generate a respective numerical representation of each word in the vocabulary; and associating each word in the vocabulary with the respective numeric representation of the word in the high-dimensional space.” and “¶ starting on Co. 7, line 19: “(27) Each of the classifiers 310 receives the numeric representation generated by the embedding function 306 and predicts a value for each field of a respective word score vector in accordance with values of a respective set of classifier parameters. Generally, each of the classifiers 310 will have different values of the classifier parameters. Each word score vector corresponds to a respective position in the sequence of words. For example, the word score vector 312 includes values for each of the predetermined set of words that is a prediction of how likely it is that the corresponding word is the word at position t-N in the sequence. The classifiers 310 can be any multiclass or multilabel classifier, e.g., a multiclass logistic regression classifier, a multiclass support vector machine classifier, a Bayesian classifier, and so on. In some implementations, instead of the classifiers 310, the concept term scoring system 300 can include ranking functions that each order the words based on the numeric representation generated by the embedding function 106, i.e., in order of predicted likelihood of being the word at the corresponding position. The ranking function may be, e.g., a hinge-loss ranking function, a pairwise ranking function, and so on.”).
Regarding claim 9, Mikolov et al. in combination with Agatsuma et al. teaches the limitations as in claim 2, above.
Mikolov et al. further teaches:
9. The information processing apparatus according to claim 2,
wherein the processor
derives a degree of similarity between the acquired item and the extracted phrase (see ¶ starting on Col. 8, line 57: “(33) Once the word prediction system 100 and the word prediction system 300 have been trained and the parameters of the embedding function 106 and the embedding function 306 have been adjusted, the numeric representations produced by the embedding functions can be used for a variety of purposes other than as input to a classifier. For example, by training the word prediction system 100 or the word prediction system 300 to generate trained values of embedding function parameters for the embedding function 106 or the embedding function 306, e.g., as described below with reference to FIG. 5, the numeric representations produced by the embedding functions can encode many useful regularities. That is, the positions of the representations in the high-dimensional space can reflect syntactic similarities, e.g., showing that, by virtue of the positions of the numerical representations of each word in the space, words that are similar to the word “small” include the words “smaller,” “tiny,” smallest,” and so on, and semantic similarities, e.g., showing that the word “queen” is similar to the words “king” and “prince.” Furthermore, because of the encoded regularities, the numeric representations may show that the word “king” is similar to the word “queen” in the same sense as the word “prince” is similar to the word “princess,” and alternatively that the word “king” is similar to the word “prince” as the word “queen” is similar to the word “princess.” Advantageously, operations can be performed on the numeric representations to identify words that have a desired relationship to other words. In particular, vector subtraction and vector addition operations performed on floating point vectors generated by an embedding function in accordance with trained values of the parameters of the embedding function can be used to determine relationships between words. For example, in order to identify a word that has a similar relationship to a word A as a word B has to a word C, the following operation may be performed on the vectors representing words A, B, and C: vector(A)−vector(B)+vector(C). For example, the operation vector(“King”)−vector(“Man”)+vector(“Woman”) may result in a vector that is closest to the vector representation of the word “Queen.”), and
performs weighting by setting a weight coefficient of the statistical value of the phrase of which the degree of similarity is equal to or larger than a certain value to a value larger than a weight coefficient of the statistical value of the phrase of which the degree of similarity is smaller than the certain value (see ¶ starting at Col. 2, line 51, ¶ starting on Co. 7, line 19, and ¶ starting on Col. 8, line 57 citations as in claim 6, above. More specifically: “¶ starting at Col. 2, line 51: “…processing each word in the vocabulary using the embedding function in accordance with the trained values of the embedding function parameters to generate a respective numerical representation of each word in the vocabulary; and associating each word in the vocabulary with the respective numeric representation of the word in the high-dimensional space.” and “¶ starting on Co. 7, line 19: “(27) Each of the classifiers 310 receives the numeric representation generated by the embedding function 306 and predicts a value for each field of a respective word score vector in accordance with values of a respective set of classifier parameters. Generally, each of the classifiers 310 will have different values of the classifier parameters. Each word score vector corresponds to a respective position in the sequence of words. For example, the word score vector 312 includes values for each of the predetermined set of words that is a prediction of how likely it is that the corresponding word is the word at position t-N in the sequence. The classifiers 310 can be any multiclass or multilabel classifier, e.g., a multiclass logistic regression classifier, a multiclass support vector machine classifier, a Bayesian classifier, and so on. In some implementations, instead of the classifiers 310, the concept term scoring system 300 can include ranking functions that each order the words based on the numeric representation generated by the embedding function 106, i.e., in order of predicted likelihood of being the word at the corresponding position. The ranking function may be, e.g., a hinge-loss ranking function, a pairwise ranking function, and so on.”).
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Keisha Y Castillo-Torres whose telephone number is (571)272-3975. The examiner can normally be reached Monday - Friday, 9:00 am - 4:00 pm (EST).
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Pierre-Louis Desir can be reached at (571)272-7799. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
Keisha Y. Castillo-Torres
Examiner
Art Unit 2659
/Keisha Y. Castillo-Torres/Examiner, Art Unit 2659
/PIERRE LOUIS DESIR/Supervisory Patent Examiner, Art Unit 2659