DETAILED ACTION
This communication is in response to the Application filed on 11/22/2024. Claims 1-12 are pending and have been examined.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 13, 2013, is being examined under the first inventor to file provisions of the AIA .
Information Disclosure Statement
The information disclosure statements (IDS) submitted on 11/22/2024, 9/19/2025, and 4/29/2026 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Priority
Applicant claims the benefit of Japan PCT Application No. JP 2022022525, filed 06/02/2022. Claims 1-12 have been afforded the benefit of this filing date.
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.
Claims 1-12 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claims 1, 5, and 9 recite A [non-transitory computer readable recording medium] storing an information processing program causing a [computer] to execute a process comprising: calculating individual vectors of a plurality of continuous sentences that have a relationship with preceding and following sentences; generating a [machine learning model] that predicts a sentence vector of a sentence input next to a certain sentence when a vector of the certain sentence is input to the machine learning model, by sequentially inputting the vectors of the plurality of sentences to the machine learning model and training the machine learning model; calculating a vector of a first sentence and a vector of a second sentence next to the first sentence; and calculating a vector of a sentence predicted to be next to the first sentence by inputting the vector of the first sentence to the machine learning model, and determining whether or not the vector of the second sentence is appropriate.
The limitations in these claims, as drafted, are a process that, under broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. The human mind is capable of converting text into vectors by assigning numerical values to words/letters within the text. A human is then capable of finding another sentence based on the vectorized version of a sentence. This could be something as simple as presenting the next sentence vector in numerical order. A human can also compare vectors and/or sentences to verify if the next predicted sentence was correct. Finally, the decision to use these sentence vectors as training data is a design decision that the human mind is capable of making. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claims recite an abstract idea.
This judicial exception is not integrated into a practical application. The claims recite the additional component of a non-transitory computer readable recording medium, a computer, and a machine learning model. The recording medium is merely used to apply the mental process via a computing device. The recording medium is detailed in paragraph 104 of the specification with generic examples provided for its implementation. The computer is merely used to apply the mental process. The computer is detailed in paragraph 100 of the specification with a generic description provided for its implementation. The machine learning model is merely used to apply the mental process via a computing device. The machine learning model is detailed in paragraph 26 of the specification with general-purpose examples provided for its implementation. Claim 9 specifically lists the additional components of a memory and a processor. The memory is merely used to apply the mental process via a computing device. The memory is detailed in paragraph 47 of the specification with generic examples provided for its implementation. The processor is merely used to apply the mental process via a computing device. The processor is detailed in paragraph 53 of the specification with generic examples provided for its implementation. Accordingly, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claims are not patent eligible.
Claims 2, 6, and 10 recite wherein the determining is determining whether or not the vector of the second sentence is appropriate based on a cosine similarity between the vector predicted by inputting the vector of the first sentence to the machine learning model and the vector of the second sentence.
The limitations in these claims, as drafted, are a process that, under broadest reasonable interpretation, covers a mathematical concept being performed by generic computer components. A human is capable of performing mathematical operations such as calculating cosine similarity of two vectors. If a claim limitation, under its broadest reasonable interpretation, covers mathematical relationships or equations but being performed by generic computer components, then it falls within the “Mathematical Concepts” grouping of abstract ideas. Accordingly, the claim recites an abstract idea.
This judicial exception is not integrated into a practical application. The claims do not recite any additional components that were not present in the independent claims. Accordingly, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claims are not patent eligible.
Claims 3, 7, and 11 recite wherein the plurality of continuous sentences are a plurality of sentences of which an arrangement order is determined based on an inductive method or a deductive method, and the generating of the machine learning model is sequentially inputting the vectors of the plurality of sentences of which the arrangement order is determined based on the inductive method or the deductive method to the machine learning model and training the machine learning model.
The limitations in these claims, as drafted, are a process that, under broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. The human mind is capable of ordering sentences using inductive or deductive reasoning. Furthermore, the design decision to order training data this way is a decision the human mind is capable of making. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claims recite an abstract idea.
This judicial exception is not integrated into a practical application. The claims do not recite any additional components that were not present in the independent claims. Accordingly, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claims are not patent eligible.
Claims 4, 8, and 12 recite wherein the computer is caused to further execute a process of calculating the vector of the sentence predicted to be next to the first sentence by inputting the vector of the first sentence to the machine learning model, and recommending an appropriate sentence based on the calculated vector of the sentence predicted to be next to the first sentence in order to search for a sentence similar to the calculated vector and present the searched sentence as a candidate for the appropriate sentence, in a case where it is determined that the vector of the second sentence is inappropriate.
The limitations in these claims, as drafted, are a process that, under broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. As mentioned previously the human mind is capable of converting sentences to vectors and calculating the similarity of those vectors. A human could do this for multiple candidate sentences and only select the sentence that is above a certain threshold of similarity. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claims recite an abstract idea.
This judicial exception is not integrated into a practical application. The claims do not recite any additional components that were not present in the independent claims. Accordingly, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. The claims are not patent eligible.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
Claims 1, 2, 4-6, 8-10, and 12 are rejected under 35 U.S.C. 103 as being unpatentable over China Patent Publication CN 11142870 A (Liu et al.) in view of Korea Patent Publication KR 102370729 B1 (Yeon).
Regarding Claims 1, 5, and 9, Liu et al. teaches A non-transitory computer readable recording medium storing an information processing program causing a computer to execute a process comprising:
(The embodiment of the invention relates to computer technology field, especially relates to a continuity judging model training method of text, text continuity determination method, and the corresponding electronic device and computer readable medium.) (Page 2, Paragraph 1).
Claim 5 present the alternative preamble An information processing method implemented by a computer, the information processing method comprising:
(The embodiment of the invention relates to computer technology field, especially relates to a continuity judging model training method of text, text continuity determination method, and the corresponding electronic device and computer readable medium.) (Page 2, Paragraph 1).
Claim 9 present the alternative preamble An information processing apparatus comprising: a memory; and a processor coupled to the memory and configured to execute a process including
(the hardware structure of FIG. 3 is the electronic device in the embodiment of the present invention, as shown in Figure 3, the electronic device may include a processor (SSP) 301, communication interface (Communications Interface) 302, a memory (memory) 303, and a communication bus 304.) (Page 10, Paragraph 11).
calculating individual vectors of a plurality of continuous sentences that have a relationship with preceding and following sentences;
(step S110: respectively obtaining the second word vector first word vector corresponding to the first text representation and corresponding to the second text representation.) (Page 3, Paragraph 11).
(In one embodiment can be selected, a program 305 for causing the processor 301 to the statement text to obtain a plurality of text pair, a plurality of words is the window size measured in two sentences in the sentence of text includes sequentially for packet processing of the adjacent sentence, to obtain a plurality of text.) (Page 12, Paragraph 8).
The system converts sequential text into word vectors.
generating a machine learning model that predicts a sentence vector of a (sentence input next to a certain sentence) (taught by Yeon) when a vector of the certain sentence is input to the machine learning model, by sequentially inputting the vectors of the plurality of sentences to the machine learning model and training the machine learning model;
(step S150: according to the first candidate sentence sequence characteristics, the second candidate sentence sequence characteristics, the preset reference relationship information of the first text and the second text, the text continuity judging model for training) (Page 4, Paragraph 1).
(processing the statement text to obtain a plurality of text, to a single text pair as unit, the plurality of text for the sequentially input text continuity judging model, respectively obtaining probability corresponding to a plurality of relationship with the plurality of text, wherein the text continuity judging model by text continuity in the first mode for realizing the judging training method for training of the model obtained according to the plurality of relationship probability, logic consistency for judging the sentence text.) (Page 12, Paragraph 6).
The model is trained using sequential text vectors. The model takes an input text and outputs a plurality of text corresponding via relationship, logic consistency, and/or judging of the sentence.
calculating a vector of a first sentence and a vector of a second sentence next to the first sentence;
(According to a first aspect of the present invention, A text training method provides continuity judging model, comprising: second word vectors respectively obtaining first word vector corresponding to the first text representation and corresponding to the second text representation; by cross-attention mechanism vector representing the first word and the second word vector representation is processed to obtain corresponding for characterizing the relationship between the first text and the second text, a plurality of different characteristic according to the plurality of different features.) (Page 2, Paragraph 5).
The text is converted into word vectors.
and calculating a vector of a sentence predicted to be next to the first sentence by inputting the vector of the first sentence to the machine learning model, and determining whether or not the vector of the second sentence is appropriate.
(For example, predicting the relation information can be relationship probability, similarly, reference relation information also can be the relation probability (The first text is on a text of the second text is 1, otherwise it is 0), then according to the prediction relationship between probability of obtaining the preset reference relationship probability, calculating the loss function corresponding to the loss value, according to the loss value of text coherence judging model for training, until satisfy the training termination condition.) (Page 9, Paragraph 7).
(to train the model by using the first text and the second text, the first text and the second text has a relationship prediction with reference to , which is characterized by reference to the relation information, it can be used for judging the training effect of labelling information. Based on this, it can judge the difference between prediction relation information with reference to the relation information for training to obtain, as calculating the loss value by the loss function, the loss value can represent the difference between the two.) (Page 4, Paragraph 7).
A loss value is calculated representing if a sentence is appropriate. The text input to the model produces an output based on the loss value and other metrics.
Liu et al. does not explicitly teach: generating a machine learning model that predicts a sentence vector of a sentence input next to a certain sentence when a vector of the certain sentence is input to the machine learning model
However, Yeon teaches generating a machine learning model that predicts a sentence vector of a sentence input next to a certain sentence when a vector of the certain sentence is input to the machine learning model
(The word/phrase sorting system 210 then outputs the sorted word and phrases 232 according to the sorting models 234 that have been generated based on the input data. Basically, in the sorting system cited above, models are trained to identify word correspondences. In the alignment technique, as shown in FIG. 6 , first, a word alignment is found between words in text segments. The system then assigns a probability to each alignment and optimizes the probability based on subsequent training data to produce a more accurate model. Outputting the aligned words and phrases 232 with the alignment models 234 is shown in block 236 of FIG. 5 .) (Page 7, Paragraph 5).
Yeon explicitly states outputting a sentence based on a sentence that was input into it whereas Liu et al. is focuses more on producing information for the input/output sentence. It does so by using vectorization and cosine similarity just like the instant application.
It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify the text analysis and production model as taught by Liu et al. to explicitly output related sentences as taught by Yeon. This would have been an obvious improvement as related sentences can reveal more ways to produce a sentence and allows the user to select a suitable one (Yeon, Page 2, Paragraph 5).
Regarding Claims 2, 6, and 10, Liu et al. in view of Yeon teaches the system of claims 1, 5, and 9.
Furthermore, Liu et al. teaches wherein the determining is determining whether or not the vector of the second sentence is appropriate based on a cosine similarity between the vector predicted by inputting the vector of the first sentence to the machine learning model and the vector of the second sentence.
(generating a resultant vector, to forecast the synthetic vector to obtain a prediction relationship information between the first word vector sequence and the second word vector sequence; The difference between the predicted relation information and the relation information between the reference , the model parameter adjusting consistency judging model of the text.) (Page 12, Paragraph 4).
(calculating the degree of matching between the "QUERY" and "VALUE" to realize the determining and extracting the characteristic between the two text from multiple angles, obtaining characterization of relationship between the first text and the second text, a plurality of different features.) (Page 4, Paragraph 5).
Liu et al. teaches calculates a degree of matching based on the angles of the vectors.
Furthermore, Yeon teaches based on a cosine similarity
(The sentence display unit 35 may calculate a similarity between example sentence vector values indicating context information of each sentence stored in the database 40 and an input sentence vector value. In this case, the sentence display unit 35 may use at least one of a Euclidean distance, a cosine similarity, and a Tanimoto coefficient as a similarity calculation method.) (Page 4, Paragraph 4).
Yeon teaches calculating similarity using a cosine similarity. The similarity is used to determine which sentences the model presents to the user.
Regarding Claims 4, 8, and 12, Liu et al. in view of Yeon teaches the system of claims 1, 5, and 9.
Furthermore, Yeon teaches wherein the computer is caused to further execute a process of calculating the vector of the sentence predicted to be next to the first sentence by inputting the vector of the first sentence to the machine learning model, and recommending an appropriate sentence based on the calculated vector of the sentence predicted to be next to the first sentence in order to search for a sentence similar to the calculated vector and present the searched sentence as a candidate for the appropriate sentence, in a case where it is determined that the vector of the second sentence is inappropriate.
(The word/phrase sorting system 210 then outputs the sorted word and phrases 232 according to the sorting models 234 that have been generated based on the input data. Basically, in the sorting system cited above, models are trained to identify word correspondences. In the alignment technique, as shown in FIG. 6 , first, a word alignment is found between words in text segments. The system then assigns a probability to each alignment and optimizes the probability based on subsequent training data to produce a more accurate model. Outputting the aligned words and phrases 232 with the alignment models 234 is shown in block 236 of FIG. 5 .) (Page 7, Paragraph 5).
(Meanwhile, the DB construction unit 33 may identify similar sentences from among the collected sentences, extract representative sentences representing the identified similar sentences, and store the extracted sentences in the database 40 . For example, among the collected sentences, sentences composed of a combination of different words may exist even if the context is the same. The DB construction unit 33 identifies similar sentences in order to reduce the load by preventing these similar sentences from being overlapped and stored in the database 40, and extracts representative sentences that can be expressed by integrating the identified similar sentences.) (Page 9, Paragraph 2).
Yeon gets a collection of sentences in a database, calculates a similarity for them, and then presents the sentences which are similar enough to the input text segment. In this sense it is performing the same behavior as the claim for sentences that were not deemed similar enough to the input.
Claims 3, 7, and 11 are rejected under 35 U.S.C. 103 as being unpatentable over China Patent Publication CN 11142870 A (Liu et al.) in view of Korea Patent Publication KR 102370729 B1 (Yeon) and further in view of China Patent Publication CN 110119446 A (Li et al.)
Regarding Claims 3, 7, and 11, Liu et al. in view of Yeon teaches the system of claims 2, 6, 10.
Furthermore, Liu et al. teaches wherein the plurality of continuous sentences are a plurality of sentences of which an arrangement order is determined based on an (inductive method or a deductive method) (Taught by Li et al.) , and the generating of the machine learning model is sequentially inputting the vectors of the plurality of sentences of which the arrangement order is determined based on the inductive method or the deductive method to the machine learning model and training the machine learning model.
(In the embodiment of the invention, the first text and the second text can be text with logical continuity, also can be the text does not have logical consistency. These text can be gained from text sample prepared in advance set for text continuity judging model for training. according to whether there is logical consistency between the first text and the second text, can be the set reference relationship information, for example, if the first text and the second text does not have logical continuity, which references the relation information is set to 0, otherwise, if the first text and the second text with logical continuity, which references the relation information is set to 1.) (Page 4, Paragraph 3).
(obtaining the first text and second text prediction relationship information, such as between the two has a probability of logical consistency, such as the first text and the second text is the probability of sentence coherent context, and so on. to train the model by using the first text and the second text) (Page 4, Paragraph 7).
Liu et al. organizes the training sentences according to logical consistency. While this could include inductive or deductive reasoning it is not explicitly stated to do so.
Liu et al. in view of Yeon does not explicitly teach: wherein the plurality of continuous sentences are a plurality of sentences of which an arrangement order is determined based on an inductive method or a deductive method
However, Li et al. teaches wherein the plurality of continuous sentences are a plurality of sentences of which an arrangement order is determined based on an inductive method or a deductive method
(body creator 220 processing and generating one or more bodies and creating effective inference rules for rule constructor 225, a predicate and similar structure. main body and the inference rules and predicate are used to generate a knowledge model … Conversely, body creator 220 for creating main body and the data associated with the main data acquisition by the data collector 215 to use. For this purpose, the body builder 220 analyzing a large amount of data and organizing the data. For example, body creator 220 deductive learning techniques may be used, such as various types of technology for organizing data and clustering the data.) (Page 8, Paragraph 5 to Page 9, Paragraph 1).
Li et al. teaches using deductive learning techniques to organize data that is used to a generate a knowledge model.
It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify the text analysis and production model as taught by Liu et al. in view of Yeon to use deductive reasoning techniques when organizing training data as taught by Yeon. This would have been an obvious improvement as implementing inference rules such as deductive logic allows artificial intelligence to more closely mimic human behavior which is relevant to generating sentences (Li et al. Page 2, Paragraphs 2-4).
Additional References
US Patent Publication US 20190156915 A1 (Zhang et al.) teaches a system which converts protein sequences into word vectors and makes predictions on the binding sites. This was included because the specification of the instant application describes a similar application for this purpose. This shows that similar systems have been taught already if the applicant were considering claiming these aspects.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NICHOLAS DANIEL LOWEN whose telephone number is (571)272-5828. The examiner can normally be reached Mon-Fri 8:00am - 4:00pm.
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, Paras D Shah can be reached at (571) 270-1650. 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.
/NICHOLAS D LOWEN/Examiner, Art Unit 2653
/Paras D Shah/Supervisory Patent Examiner, Art Unit 2653
07/02/2026
Prosecution Insights