DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
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.
Claim(s) 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Arunachalam et al (20250211549) in view of Mane et al (20240095446).
As per claim 1, Arunachalam et al (20250211549) teaches a method for data processing, comprising:
converting a set of metadata from a first structured format to a second serialized format, wherein the set of metadata corresponds to a data object within a data store (as, taking the metadata tags that include project identifier, title, chunk identifier, and the like – para 0079, and para 0015; );
generating a first natural language summary corresponding to the data object by inputting the set of metadata in the second serialized format into a LLM (as, generating a summary from a LLM -- see para 0071, the LLM generates a summary from the metadata pointers, and processing the metadata associated chunks until a summary is generated) ;
generating a second natural language summary corresponding to the data object by inputting a natural language query for the data object into the LLM (generating a second natural language summary of the project transcripts – para 0019);
and causing for display an indication of the data object as being related to the natural language query (as displaying, in a session, the created answer/summary an compared to the original documents/transcripts – para 0076) based at least in part on vector-space comparison of a vectorized version of the first natural language summary and a vectorized version of the second natural language summary (which is based on the vector comparison in para 0066 -- as using a similarity calculation to measure the distance between the two vectors – para 0066; a cosine similarity search is used to find/compare the similarity of the 2 vectors – end of para 0066; examiner notes that one vector is generated/represents the information request – para 0066, first 3 sentences; and the second vector is a representation of the project data – para 0066; examiner notes that the information request is represented by the summary from metadata – see para 0014, into para 0015 , and the second vector project data represents the natural language processing – see para 0015 – documents/transcripts with the data of a project).
As per claim 1, Arunachalam et al (20250211549) operates on metadata, and text items, but does not specify that the second serialized data is in an unstructured data format; Mane et al (20240095446) teaches the concept of extracting metadata and meaning from a data stream, then passing the remaining unstructured text to a natural language processing unit (in other words, the mix of metadata and ‘unstructured’ text, is, by definition, ‘having structure’ since the metadata is marked as formatted data, and the remaining text being unstructured; after the metadata is processed, clearly, only the remaining unstructured text is sent to the natural language processor – see para 0018 ). Therefore, it would have been obvious to one of ordinary skill in the art of ML text processing to further define the process in Arunachalam et al (20250211549) with processing metadata and further sending the remaining unstructured text to a NLP, as taught by Mane et al (20240095446), because it would advantageously generate the text into a standardized structure for further input, as needed (see end of para 0018, Mane et al (20240095446)).
As per claim 2, the combination of Arunachalam et al (20250211549) in view of Mane et al (20240095446) teaches the method of claim 1, further comprising: generating the vectorized version of the first natural language summary and the vectorized version of the second natural language summary using an embedding model (see Arunachalam et al (20250211549), as, using embedding LLM to generate vectors from the data chunks, which are embedding in the embedding space for the vectors – para 0080; the ‘embedding chunks’ can be from the metadata tags/info – para 0079 – which is the first language summary – see above; or the transcript from which the chunks were generated – para 0079; and see para 0071, showing the summary generated from the chunks).
As per claim 3, the combination of Arunachalam et al (20250211549) in view of Mane et al (20240095446) teaches the method of claim 1, wherein causing for display the indication of the data object further comprises:
performing the vector-space comparison based at least in part on measuring a distance between the vectorized version of the first natural language summary and the vectorized version of the second natural language summary (see Arunachalam et al (20250211549), as, using a similarity calculation to measure the distance between the two vectors – para 0066; a cosine similarity search is used to find/compare the similarity of the 2 vectors – end of para 0066; examiner notes that one vector is generated/represents the information request – para 0066, first 3 sentences; and the second vector is a representation of the project data – para 0066; examiner notes that the information request is represented by the summary from metadata – see para 0014, into para 0015 , and the second vector project data represents the natural language processing – see para 0015 – documents/transcripts with the data of a project).
As per claim 4, the combination of Arunachalam et al (20250211549) in view of Mane et al (20240095446) teaches the method of claim 3, wherein performing the vector-space comparison further comprises: performing a ranking procedure to rank a plurality of vector distances (see Arunachalam et al (20250211549), as, para 0066, determining the vectors with the shortest distance; examiner notes that by definition, to track a ‘shortest’ distance, is to calculate each vector match and then compare each distance measure with each other, to determine which distance is ‘shortest’ – see also, “top K most similar vectors”, in para 0066).
As per claim 5, the combination of Arunachalam et al (20250211549) in view of Mane et al (20240095446) teaches the method of claim 1, wherein generating the first natural language summary further comprises:
generating a prompt indicating that the set of metadata is in the second serialized format for the LLM, wherein the first natural language summary is generated in accordance with the prompt (see Arunachalam et al (20250211549), as, the generative AI system determines, from the request of information, into predefined prompts to the LLM, to generate a more complete response – para 0034), and wherein the second serialized format is an unstructured string of characters (see Mane et al (20240095446), para 0018).
As per claim 6, the combination of Arunachalam et al (20250211549) in view of Mane et al (20240095446) teaches the method of claim 1, further comprising:
storing the vectorized version of the first natural language summary in a vector database (see Arunachalam et al (20250211549), as, storing the aforementioned vectors in a database/dictionary – para 0057).
As per claim 7, the combination of Arunachalam et al (20250211549) in view of Mane et al (20240095446) teaches the method of claim 1, wherein the second natural language summary corresponds to a hypothetical data object related to the natural language query (see Arunachalam et al (20250211549), as, the second natural language summary ties the transcript to a possible data object -- see para 0013, wherein the transcript may be labeled as tables/images meeting notes, etc.; also see para 0015, wherein the transcript sections can be labeled (ie, data object) by title, hierarchy, data source, etc.).
As per claim 8, the combination of Arunachalam et al (20250211549) in view of Mane et al (20240095446) teaches the method of claim 1, wherein the set of metadata in the first structured format indicates a plurality of attributes associated with the data object (see Arunachalam et al (20250211549), as, the metadata indicates a plurality of attributes, such as title, document source, hierarchy, etc. – para 0015).
As per claim 9, the combination of Arunachalam et al (20250211549) in view of Mane et al (20240095446) teaches the method of claim 8, wherein generating the first natural language summary is based at least in part on the plurality of attributes (see Arunachalam et al (20250211549), as, using the attributes, in the metadata tags, to identify chunks (para 0018, first 2 sentences) that are used by the generative AI system to generate a summary – para 0018, last sentence).
As per claim 10, the combination of Arunachalam et al (20250211549) in view of Mane et al (20240095446) teaches the method of claim 1, wherein the data object comprises structured data in tabular form – see Arunachalam et al (20250211549), as, para 0015, wherein the data objects can be title, document source; see para 0038, wherein the tag can also include subtitle, hierarchy, etc. – examiner notes, that such information is well known in the art to be separated by delimiters and hence, in tabular form).
Claims 11-19 are apparatus claims that perform the steps found in method claims 1-10 above and as such, claims 11-19 are similar in scope and content to claims 1-10 above; therefore, claims 11-19 are rejected under similar rationale as presented against claims 1-10 above. Furthermore, Arunachalam et al (20250211549) teaches processors executing the steps (para 0101) stored in system memories (para 0103).
Claim 20 is a non-transitory computer readable medium storing executable instructions when executed by a processor, performs the steps found throughout method claims 1-10 above and as such, claim 20 is similar in scope and content to claims 1-10 above; therefore, claim 20 is rejected under similar rationale as presented against claims 1-10 above. Furthermore, Arunachalam et al (20250211549) teaches storage devices storing the executable code to perform the disclosed steps (para 0103).
Response to Arguments
Applicant’s arguments with respect to claim(s) have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. Applicants amendment to overcome the claim objections, are accepted, and the claim objections are removed. Applicant’s arguments, on pp 7-8 of the response, are toward the amended claim language; examiner notes the introduction of the Mane et al (20240095446) reference, to further define the data structure entering into the NLP – ie, Mane et al (20240095446) teaches the interpretation of the metadata (not unstructured text), and then submitting the unstructured text only, with the intent/interpretation of the metadata, to assist in the recognition of the input. See mappings above.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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.
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Please see related art listed on the PTO-892 form.
Furthermore, the following references were found to teach certain aspects of applicants specification and/or claim features:
Toward the mix of unstructured/structured data with metadata, and the output ‘cleaned’ with unstructured data into further language processing:
Ward et al (20230385559), para 0118.
Cenciotti et al et al (20220207038), para 0038
Toward other claim/spec features:
Cho et al (20220414338) teaches the processing of transcripts, separately, based on metadata, annotated transcripts, and unassisted conversion transcripts – para 0109-0116
Mukherjee et al (20250124295) teaches machine learning training based on transcripts, leveraged transcripts and text sequences – para 0027, figure 4.
Haikin et al (20250165720) teaches analytics techniques that multi-parallel analysis of transcripts of the interaction/event, metadata of the interaction/event, and application/interaction path of the event – see para 0045, see Figure 6 reflecting back on Figure 4.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Michael Opsasnick, telephone number (571)272-7623, who is available Monday-Friday, 9am-5pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, Mr. Richemond Dorvil, can be reached at (571)272-7602. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
/Michael N Opsasnick/Primary Examiner, Art Unit 2658 05/26/2026
Prosecution Insights