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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 12 March 2026 has been entered.
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 22-24 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a mental process without significantly more.
Independent claim 22 is directed to:
“An optimization method for an input search formula of a patent literature performed by a server, comprising:
identifying, in a user interface, the input search formula for the patent literature, the input search formula including a plurality of first search words;
classifying, by the server, the input search formula into a plurality of groups based on a preset search operator;
adding, by the server, to each of the groups, at least one second search word associated with at least one first search word included in a respective group by using a first search operator;
for each of the groups after the adding of the at least one second search word, extracting, by the server, from among a plurality of data items of patent literature, a data item having the highest probability of being searched for the respective group, based on a model trained using original patent literature text data and translated patent literature text data for the plurality of data items of patent literature;
adding, by the server, for each of the groups, an item field operator corresponding to the extracted data item;
generating, by the server, a final search formula by connecting the plurality of groups with a second search operator; and
displaying, by the server, the generated final search formula,
wherein the highest probability is determined based on at least one of:
(i) whether the at least one second search word is a synonym of at least one of the first search words,
(ii) whether the at least one second search word is translated from at least one of the first search words, and
(iii) whether the at least one second search word exists in the same patent literature as at least one of the first search words,
wherein the server provides the at least one second search word for a selected group according to a group-selection input and outputs each second search word as a selectable icon, and
wherein, in response to receiving a selection input for one of the selectable icons, the server activates the selected icon and displays information associated with the activated icon in a window of the selected group.”
This is a mental process because the identifying, classifying, adding, adding, generating, and data definition steps can be performed by a human being with a generic computer.
It is additionally noted that the specification explicitly says that the goal of the invention is to “provide a higher-quality patent literature search service by optimizing and expanding a patent literature search formula input by a user to a high-quality patent literature search formula as if created by an expert” (see specification as filed, paragraph [0006]). Converting a search query into one designed by a human expert is something that a human would be capable of with pen and paper or a generic computer.
The additional elements in the claim beyond a mental process are “a server,” “a user interface,” “extracting by the server,” “displaying,” “and a selectable icon.”
This judicial exception is not integrated into a practical application because the claimed additional elements do not appear to improve the processing of a computer, require the use of a specific machine, effect a transformation or reduction of a particular article to a different state or thing, or provide a technological solution to a technological problem.
The additional element of the server is recited at a high level of generality. The server appears to be a generic computing hardware elements. The recitation of generic hardware is little more than using a computer to perform an abstract idea, see MPEP 2106.05(f)(2). The additional elements of “a user interface,” “displaying…,” and “a selectable icon” are all directed towards generic display of analyzed data. Displaying an output of a data analysis is insignificant extra-solution activity and is well known (see MPEP 2106.05(g)((3)). Displaying results of a data analysis also does not show an improvement to technology (see MPEP 2106.05(a)(II)). The additional element of “extracting data” appears to be a data gathering step, and is thus mere pre-solution insignificant activity (see MPEP 2106.05(g).
It is noted that none of the additional elements appear to improve the processing of a computer, require the use of a specific machine, effect a transformation or reduction of a particular article to a different state or thing, or provide a technological solution to a technological problem. As such, none of the additional elements appear to integrate the judicial exception into a practical application.
None of the additional elements are sufficient to amount to significantly more than the judicial exception, in part or in whole.
The “server” is little more than using a computer to perform an abstract idea, see MPEP 2106.05(f)(2). The additional elements of “a user interface,” “displaying…,” and “a selectable icon” are all generic user interface elements, insignificant extra-solution activity, and are well known (see MPEP 2106.05(g)((3). Extracting data is merely extra-solution activity data gathering and is well understood, routine, and conventional (see MPEP 2106.05(g)).
Dependent claims 23 and 24 merely add additional data analysis steps. Because only additional data analysis steps are claimed, without additional elements, these claims are directed towards mental processes. Mental processes, on their own, cannot embody a practical application nor be sufficient to amount to significantly more than a mental process. In other words, a more efficient mental process is still a mental process. As such, dependent claims 23 and 24 are similarly rejected under 35 USC 101 as being directed towards a mental process.
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.
Claims 22-23 are rejected under 35 U.S.C. 103 as being unpatentable over Miller et al. (US Pre-Grant Publication 2017/0235799) in view of Yadav et al. (US Pre-Grant Publication 2021/0019309), and further in view of Huber et al. (US Pre-Grant Publication 2020/0273493).
As to claim 22, Miller teaches an optimization method for an input search formula of a patent literature performed by a server, comprising:
identifying, in a user interface, the input search formula for the patent literature, the input search formula including a plurality of first search words (see Miller Figure 10 and paragraph [0069]. A query has been constructed by a user. Figure 10 shows an intermediary stage of the query construction process. The query is built in the canvas UI of element 310 as a collection of concept stacks containing concept nodes, see Figure 3 and paragraph [0042]. The query is written in textual form in element 360. Also see paragraph [0026], which discusses how the context of Miller is to search patent literature);
classifying, by the server, the input search formula into a plurality of groups based on a preset search operator (see Miller paragraphs [0042]-[0043] and [0069] and Figures 3 and 10. Each of the query elements is classified into different ‘concept stacks.’ Each of the concept groups may be joined together with conjunctive operators);
adding, by the server, to each of the groups, at least one second search word associated with at least one first search word included in a respective group by using a first search operator (see Miller paragraphs [0069]-[0070] and Figure 10. Each concept stack has a “suggestions” dropdown box, element 343. By selecting the “suggestions” box, similar search words related to an original search word are shown. When the user selects one of the suggested query terms, the query search term is added to the original search terms with an OR operator, see paragraph [0070]);
…
generating, by the server, a final search formula by connecting the plurality of groups with a second search operator (see Figure 10 and paragraphs [0069]-[0070]. The final search formula is shown in element 360. As noted in element 360 and in paragraph [0070], the final search formula connects and combines the query elements from each concept stacks with additional search operators, such as “AND” or “PARAGRAPH”); and
displaying, by the server, the generated final search formula (see Figure 10 and paragraphs [0069]-[0070]),
…
wherein the server provides the at least one second search word for a selected group according to a group-selection input and outputs each second search word as a selectable icon (see Miller paragraphs [0070] and Figure 10. An additional second word results from a selection of a user of a suggested term. This is output as a node (an icon) and added to the query representation in the same way as element 342 of Figure 10), and
wherein, in response to receiving a selection input for one of the selectable icons, the server activates the selected icon and displays information associated with the activated icon in a window of the selected group (see Figure 10 and paragraph [0070]. When a suggested input word is selected, the input word is added as an icon. This icon is “activated” as part of the search. Additionally, the word of the icon is displayed in the search window as a part of the selected group).
Miller does not explicitly teach:
for each of the groups after the adding of the at least one second search word, extracting, by the server, from among a plurality of data items of patent literature, a data item having the highest probability of being searched for the respective group, based on a model trained using original patent literature text data and translated patent literature text data for the plurality of data items of patent literature;
adding, by the server, for each of the groups, an item field operator corresponding to the extracted data item;
wherein the highest probability is determined based on at least one of:
(i) whether the at least one second search word is a synonym of at least one of the first search words,
(ii) whether the at least one second search word is translated from at least one of the first search words, and
(iii) whether the at least one second search word exists in the same patent literature as at least one of the first search words,
Yadav teaches:
for each of the groups after the adding of the at least one second search word, extracting, by the server, from among a plurality of data items of patent literature, a data item having [a probability of] being searched for the respective group (see Yadav paragraph [0322] and [0305]. A probability score associated with a column of a database may be made based on terms within a query and potentially related columns. Also see paragraphs [0339] and [0345]-[0346]. As noted above, the columns are based on the underlying data being analyzed. While Applicant claimed “patent literature,” it is noted that no particular value of any sort of “patent literature” has any functional effect on the limitations. Thus, the limitation of “patent literature” is obvious as being merely a design choice of non-functional descriptive material that one of ordinary skill in the art would recognize as a data type that the claims could be operated with. Also see MPEP 2111.05(III), which discusses the difference between functional and non-functional descriptive material),
based on a model trained using original patent literature text data and translated patent literature text data for the plurality of data items of patent literature (see Yadav paragraph [0322] and [0305] and [0345]-[0346]. Data items may be added to a query based on probability. It is noted that any particular data types limited to the field of “patent literature” have no functional bearing on the claim. It is merely a type of data that the method is being applied to. Thus, because training based on “original patent literature” and “translated patent literature” is merely a type of data and no specific values of this data are part of any claimed calculations, this limitation is obvious. It is noted that the translation occurs outside the context of the claim. Thus, the translated patent literature of the claim is merely a data input);
adding, by the server, for each of the groups, an item field operator corresponding to the extracted data item (see Yadav paragraph [0322] for an example. A data item, or field, may be added to a search query. This may include, as in the example shown in paragraph [0322], adding in a “genre=documentary” data item to the query);
It would have been obvious to one of ordinary skill in the art before the earliest filing date of the invention to have modified Miller by the teachings of Yadav because both references are directed towards querying datasets and inferring or guessing user intent. Yadav merely adds additional metrics that may be used to determine a user intent, which will increase the ability of Miller to find relevant results.
Huber teaches:
Extracting, by the server, from among a plurality of data items of patent literature, a data item having the highest probability of being searched for the respective group, based on a model trained using original patent literature text data and translated patent literature text data for the plurality of data items of patent literature (see Huber paragraphs [0063]-[0064]. Huber uses a machine learning model to extract terms from a transcript and identify terms with the highest probability of being searched. Huber then recommends those terms to the user for a query. It is noted that training based on “original patent literature text data and translated patent literature text data” appears merely to be a type data type. Thus, because training based on “patent literature” is merely a type of data and no specific values of this data are part of any claimed calculations or analysis, this data type limitation is obvious);
wherein the highest probability is determined based on at least one of:
(i) whether the at least one second search word is a synonym of at least one of the first search words (see Huber paragraphs [0063]-[0064]. Huber considers synonyms when identifying terms with the highest probability),
(ii) whether the at least one second search word is translated from at least one of the first search words, and
(iii) whether the at least one second search word exists in the same patent literature as at least one of the first search words.
It would have been obvious to one of ordinary skill in the art before the earliest filing date of the invention to have modified Miller by the teachings of Huber because both references are directed towards querying datasets and discerning user intent. Huber merely adds an additional method that may be used to determine a user intent, which will increase the ability of Miller to find relevant results.
As to claim 23, Miller as modified by Yadav teaches the optimization method of claim 22, wherein the step of extracting the data item having the highest probability of searching the each group includes:
a step of deriving an average amount for each data item (see Yadav paragraph [0064]. Usage data may be considered, wherein the usage data is a derived average amount for query tokens. Also see paragraph [0488] for determining statistics around word usage in a query. Note that Huber is relied upon to teach identifying words having a ‘highest’ probability); and
a step of extracting the data item having the highest probability of being searched by calculating an average number of times the each group is searched in the each data item compared to a derived average amount of each data item (see Yadav paragraphs [0064] for tracking average word usage. Paragraph [0484] shows how this may result in an inference of a column being selected over another column. Paragraph [0488] shows comparing a frequency of search for each group in response to a token. Note that Huber is relied upon to teach identifying words having a ‘highest’ probability).
Claim 24 is rejected under 35 U.S.C. 103 as being unpatentable over Miller et al. (US Pre-Grant Publication 2017/0235799) in view of Yadav et al. (US Pre-Grant Publication 2021/0019309), further in view of Huber et al. (US Pre-Grant Publication 2020/0273493), and further in view of Akiyama et al. (US Pre-Grant Publication 2009/0150361).
As to claim 24, Miller as modified teaches the optimization method of claim 22.
Miller as modified does not teach wherein in a case where there are groups in which the item field operator overlaps among the plurality of groups, the step of adding the item field operator includes:
a step of connecting groups having the same item field operator with a preset search operator; and
a step of adding the same item field operator to the connected groups.
Akiyama teaches:
wherein in a case where there are groups in which the item field operator overlaps among the plurality of groups, the step of adding the item field operator includes:
a step of connecting groups having the same item field operator with a preset search operator (see Akiyama paragraph [0046]. Search items may be recognized as belonging to the same category); and
a step of adding the same item field operator to the connected groups (see Akiyama paragraph [0046]. Search objects belonging to the same category are connected with an OR operator).
It would have been obvious to one of ordinary skill in the art before the earliest filing date of the invention to have modified Miller by the teachings of Akiyama because both references are directed towards querying datasets. Akiyama merely adds additional metrics that may be used to enhance a query by grouping objects together, improving the ability of Miller to build queries based on a user intent.
Response to Arguments
Applicant's arguments filed 12 March 2026 have been fully considered but they are not persuasive.
35 USC 101 Response
Applicant argues that “claim 22 is not directed merely to a mental process or to insignificant extra- solution activity. Rather, claim 22 recites a specific server-implemented process for automatically generating a machine-usable search formula for patent literature.”
Applicant continues, arguing that “More specifically, the claimed server:
(1) classifies an input search formula into a plurality of groups based on a preset search operator;
(2) adds associated second search words to each group using a first search operator;
(3) extracts, for each group, a data item having the highest probability of being searched based on a model trained using original patent literature text data and translated patent literature text data for a plurality of patent-literature data items;
(4) adds, for each group, an item field operator corresponding to the extracted data item; and
(5) generates a final search formula by connecting the plurality of groups with a second search operator.”
Applicant concludes by arguing that “These operations are not practically performable in the human mind in the manner claimed. In particular, the claimed server applies a trained model to patent-literature data and translated patent-literature data to determine, on a group-by-group basis, which data item has the highest probability of being searched, and then automatically restructures the search formula itself by inserting corresponding item field operators.”
In response to this argument, each of the listed steps (“classifying…”, “adding…”, “extracting…”, “adding…”, and “generating a final search formula by connecting the plurality of groups with a second search operation;”) are mental process steps. Each of the listed step is merely a data analysis step. A human being equipped with a pen and paper or a generic computer is capable of each of these steps.
While the claims contain a “server,” as noted in MPEP 2106.04(a)(2) III C, “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").”
MPEP 2106.04(a)(2) III C 1-3 further elaborate on the idea that a claim may still be directed towards an abstract idea despite the use of a generic machine. Thus, though the claims may not be wholly performed in a human mind, the “classifying…”, “adding…”, “extracting…”, “adding…”, and “generating…” steps may be performed by a human with a generic computer.
The claim model is recited as merely a black box that accepts input to train (“original patent literature text data” and “translated patent literature text data”) and produces an output (“a data item having the highest probability of being searched for the respective group”). No inner workings regarding the machine learning model is claimed. It appears to simply be another data analysis element.
As noted in Recentive Ananlytics Inc v. Fox Corp, the addition of a generic machine learning model to the claims in a new data environment, without disclosing improvements to the machine learning model applied, does not integrate a mental process into a practical application. Simply applying a machine learning model to a new data environment does not render a claim patent eligible in view of 35 USC 101.
Applicant continues, arguing that “The claim is also integrated into a practical application. The invention improves computerized patent-literature retrieval by generating a search formula that is specifically adapted for execution against patent-literature data items. The claimed process does not merely present information to a user; rather, it changes the structure of the search formula in a way that controls how the patent-literature search is performed by the computer system.”
In response to this argument, “computerized patent-literature retrieval” is, generally, a method of searching data. Searching data is a mental process directed towards data analysis. A human being equipped with pen and paper or a generic machine is capable of searching data and performing data analysis. Even if the claims represent an improvement to “computerized patent-literature retrieval,” an improved mental process is still a mental process and patent ineligible.
Applicant has not identified how the additional elements of the claims, as outlined in the rejection above, integrate the claim into a practical application or are, in part or as a whole, significantly more than a mental process with citations to the specification showing such an improvement to a computer or technological field. Because Applicant has not shown this, Applicant’s argument is unpersuasive.
35 USC 103 Response
Applicant argues that “However, Yadav does not teach or suggest the claimed patent-literature-specific processing, the use of original patent literature text data together with translated patent literature text data to train the model, or the recited group-specific workflow in which a data item is extracted for each group and a corresponding item field operator is then added for that same group.”
In response to this argument, it is noted that Yadav is relied upon to show a data item extracted for a group and a corresponding item field operator is then added for that same group (see Yadav paragraph [0322] and [0305]. A probability score associated with a column of a database may be made based on terms within a query and potentially related columns. Also see paragraphs [0339] and [0345]-[0346]).
It is noted that simply performing the same operation on multiple groups is a mere duplication of parts and is obvious. As noted in MPEP 2144.04(VI)(B), “duplication of parts has no patentable significant unless a new and unexpected result is produced.”
In reHarza, 274 F.2d 669, 124 USPQ 378 (CCPA 1960) (Claims at issue were directed to a water-tight masonry structure wherein a water seal of flexible material fills the joints which form between adjacent pours of concrete. The claimed water seal has a “web” which lies in the joint, and a plurality of “ribs” projecting outwardly from each side of the web into one of the adjacent concrete slabs. The prior art disclosed a flexible water stop for preventing passage of water between masses of concrete in the shape of a plus sign (+). Although the reference did not disclose a plurality of ribs, the court held that mere duplication of parts has no patentable significance unless a new and unexpected result is produced.).
Regarding the “claimed patent-literature-specific processing, the use of original patent literature text data together with translated patent literature text data to train the model,” Examiner notes that this appears to just be a type of data input into the machine learning model. There are no limitations directed towards how particular values of the data specific to patent literature affect the model’s training or output. The data is not translated within the scope of the claim, but rather is pre-translated input data. As such, it appears to be applying the claimed functional steps merely to a specific type of data and training a model with a specific type of data. The claimed limitation of “patent literature text data” represents merely a design choice of non-functional descriptive material and would be obvious to one of ordinary skill in the art utilizing the functional aspects of the claims. Also see MPEP 2111.05(III), which discusses the difference between functional and non-functional descriptive material.
Applicant elaborates, arguing that “the present invention requires a learned model trained on patent literature text data (original and/or translated) to classify and extract a data item with the highest probability of relevance to each group. This probabilistic extraction and classification mechanism is absent from Yadav.”
In response to this argument, it is noted that newly cited reference Huber is now relied upon to show wherein a highest probability of relevance data item may be added to a search group.
As noted above, performing on operation on multiple groups is merely a duplication of parts. Yadav and Huber both show recommending a data item to an individual group. Thus, in view of Yadav and Huber and MPEP 2144.04(VI)(B), it would have been obvious to perform such an operation on multiple groups.
Also noted above training a model on “original and/or translated data” of patent literature is merely a type of data being input into a machine learning model and analyzed in a black box. Any values contained within the data particular to “patent literature” have no claimed effect on any functional step of the claims. Because of this, the type of data is a mere design choice of non-functional descriptive material and would have been obvious to one of ordinary skill in the art. If Applicant wishes for this element to receive patentable weight, Applicant should claim how particular values present in the field of patent literature data affect the machine learning model or the extraction step.
It is additionally noted that the only “classification” operation of the claims is separate from the probabilistic extraction and is taught by Miller.
Applicant continues, arguing that the UI elements of Yadav “do not disclose or imply the claimed probability-based determination of a data item for each group. The claimed invention explicitly requires a model-driven probability ranking and extraction process, which is fundamentally different from Yadav’s deterministic token suggestion.”
In response to this argument, it is noted that Huber shows a model-driven probability ranking and extraction process for suggesting objects to add to a search formula.
Applicant argues that “Yadav does not disclose any mechanism of associating an item field operator with a probabilistically extracted data item per group.”
In response to this argument, it is noted that Yadav does show a mechanism of associating an item field operator with a probabilistically inferred data item to a group (see Yadav paragraph [0322] for an example. A data item, or field, may be added to a search query. This may include, as in the example shown in paragraph [0322], adding in a “genre=documentary” data item to the query).
It is noted that performing this operation on each group is merely a duplication of parts. As noted in MPEP 2144.04(VI)(B), cited above, “duplication of parts has no patentable significant unless a new and unexpected result is produced.”
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CHARLES D ADAMS whose telephone number is (571)272-3938. The examiner can normally be reached M-F, 9-5:30 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, Neveen Abel-Jalil can be reached at 571-270-0474. 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.
/CHARLES D ADAMS/Primary Examiner, Art Unit 2152