DETAILED ACTION
1. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
2. This action is in response to amendment filed 7/31/2026, in which claims 1, 10 – 11, and 16 was amended, claims 15 and 21 was cancelled, and claims 1 – 11, 13 – 14, 16 – 20, and 22 was presented for further examination.
3. Claims 1 – 11, 13 – 14, 16 – 20, and 22 are now pending in the application.
Continued Examination Under 37 CFR 1.114
4. 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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 7/31/2026 has been entered.
Response to Arguments
5. Applicant’s arguments with respect to claims 1 – 11, 13 – 14, 16 – 20, and 22 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.
Remarks
6. As per claim 1, applicant argues in substance in pages 9 – 11 that Mondlock et al (US 12,039,263 B1), Gupta et al (IN 202121000983 A), and Bayless et al (US 2025/0112878 A1) does not disclose displaying, via the chatbot interface, a retrieved context regarding the document of the documents, wherein: the retrieved context comprises information enabling a user to verify a source of the answer; and the chatbot interface comprises a graphical user interface configured to display the retrieved context in a context retrieval drop-down menu.
In response to applicants’ argument, Examiner respectfully responds that the combine teaching of Mondlock et al (US 12,039,263 B1), Gupta et al (IN 202121000983 A), and Courcelle et al (US 2025/0293998 A1) specifically disclose displaying, via the chatbot interface, a retrieved context regarding the document of the documents, wherein: the retrieved context comprises information enabling a user to verify a source of the answer; and the chatbot interface comprises a graphical user interface configured to display the retrieved context in a context retrieval drop-down menu (para.[0032], para.[0090], para.[0105], and para.[0127]).
Courcelle discloses chatbot’s response interface includes a list of results and a visualization representing the results (see para.[0031] and para.[0033]). The chatbot receives user input and generates a response using AI/ML models . The response includes a summary of results (see para.[0081]). The user interface displays the response. The interface includes a document viewer (see para.[0090]). The user inputs a follow-up query into the chatbot to retrieve the relevant search results. The follow-up query may restrict the search to the initial search results (see para.[0007] and para.[0032]).
Thus, the rejection is maintained.
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.
7. Claims 1 – 10 and 22 are rejected under 35 U.S.C. 103 as being unpatentable over Mondlock et al (US 12,039,263 B1), in view of Gupta et al (IN 202121000983 A), in view of Courcelle et al (US 2025/0293998 A1).
As per claim 1, Mondlock et al (US 12,039,263 B1) discloses,
A method for using generative artificial intelligence to generate an answer in response to a natural language query that is directed to oil and gas exploration, drilling, and/or production (col.1 lines 21- 23; “user queries with relevant supplied data to enable LLMs to provide improved responses” and col.1 lines 34 – 37; “receiving a user query, fetching relevant external data, and submitting prompts to cause the LLM to answer the user query based on provided relevant external data”, where external data may be interpreted as “oil and gas exploration drilling, and/or production” as claimed, however, Gupta, secondary reference address query an oilfield document).
the method comprising: receiving a plurality of documents (col.6 lines 25 – 28; “retrieving copies of and/or scraping text or data from the document collections 162 and
asset collections 164 located in one or more external data sources”).
splitting the documents into chunks (col.6 lines 33 – 34; “splitting documents or assets into chunks”).
generating a plurality of embeddings based upon the chunks (col.6 lines 33 – 34; “splitting documents or assets into chunks and generating embeddings of those chunks” and col.6 lines 49 – 50; “generating embeddings from each text chunk and/or data chunk”).
storing the chunks and the embeddings in a vector database (col.6 lines 60 – 63; “document/asset/expert module 122 may save the embeddings into embeddings 168 in the external data sources 160. The embeddings 168 may comprise a vector database”).
receiving a natural language query directed to oil and gas exploration, drilling, and/or production (col.7 lines 10 – 12; “receive the query or the query plus retrieved chat history and classify the query into one or more of a plurality of pre-defined intents”).
generating a query embedding based upon the natural language query (col.7 lines 15 – 18; “generating an embedding of each predefined intent; (2) generating an embedding of the user query”).
retrieving a subset of the chunks based upon the query embedding (col.8 lines 13 – 15; “generating an embedding of the user query; and (2) compare the user query embedding to the document and asset embeddings in embeddings” and col.8 lines 41 – 44; “compare the embedding of the user query to embeddings of the text chunks and/or data chunks to identify relevant text chunks and/or relevant data chunks”).
generating an answer in response to the natural language query (col.8 lines 17 – 21; “relevant information identification module 130 may transmit the query embedding, the document embeddings, and/or the data 20 embeddings, via the LLM interface module 132, to the LLM Service” and col.8 lines 25 – 32; “topic modeling may include (1) performing topic modeling on the user query to identify one or more topic keywords; and (2) searching the document collections 144, asset collections 146, expert collections 148, document collections 162, asset collections 164, and/or expert collections 166 with the topic keywords to identify relevant documents, assets, and/or experts”).
wherein the answer is based upon the natural language query and the subset of the chunks (col.8 lines 50 – 57; “the LLM interface module 132 may transmit prompts to and receive answers from the LLM service ………… receive relevant information from the relevant text chunk and/or relevant data chunk” and col.16 lines 19 – 21; “prompts may cause the LLM service 170 to output relevant information for each of
the text chunks and data chunks”).
Mondlock does not specifically disclose oil and gas exploration, drilling, and/or production.
However, Gupta et al (IN 202121000983 A) in an analogous art discloses,
natural language query that is directed to oil and gas exploration, drilling, and/or production (para.[0004]; “file comprising an unstructured oilfield document”, para.[0015]; “various type of oilfield documents including reports”, para.[0017]; “oilfield documents generated through the various exploration and production (E&P) operations of an oilfield”, para.[0037]; “the document content type may be drilling log, production log, daily production report, well completion report, geochemistry lab report, seismic report, or other document content type”, and para.[0089]; “user, or software application, may submit a statement or query into the DBMS ……… perform computations to respond to the query. The DBMS may return the result(s) to the user or software application”).
receiving a natural language query directed to oil and gas exploration, drilling, and/or production (para.[0037]; “the document content type may be drilling log, production log, daily production report, well completion report, geochemistry lab report, seismic report, or other document content type”, and para.[0089]; ““user, or software application, may submit a statement or query into the DBMS ……… perform computations to respond to the query. The DBMS may return the result(s) to the user or software application”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate classification of unstructured oilfield document of the system of Gupta into generative AI pipelines that enable parallel large language model
(LLM) queries of the system of Mondlock to extract and arrange relevant unstructured oilfield data, thereby providing data that is easy to interpret by the user.
Neither Mondlock nor Gupta specifically disclose the generating the answer comprises generating a chatbot interface; the natural language query is a first query; the method further comprises receiving, via the chatbot interface, a second query regarding a document of the documents; and the answer comprises a summary of the subset of the chunks, the summary comprising a non-verbatim summary of a paragraph of the document of the documents; and displaying, via the chatbot interface, a retrieved context regarding the document of the documents, wherein: the retrieved context comprises information enabling a user to verify a source of the answer; and the chatbot interface comprises a graphical user interface configured to display the retrieved context in a context retrieval drop-down menu.
However, Courcelle et al (US 2025/01293998 A1) in an analogous art discloses,
the generating the answer comprises generating a chatbot interface (para.[0084]; “generates and sends a response 182 as the answer of the chatbot to the user prompt 170. The response 182 is then displayed on the user interface”).
the natural language query is a first query (para.[0061]; “user 105c interacts with the
chatbot by submitting the user prompt 170, such as a natural language question or statement” and para.[0093]; “a user prompt such as a natural language query is received”).
the method further comprises receiving, via the chatbot interface, a second query regarding a document of the documents (para.[0022]; “ask follow-up questions about that conversation result” and para.[0028]; “the user can immediately ask a follow-up question, and the chatbot can use the information from the previous response (including
potential intermediate information or information derived from multiple prior interactions) to inform the creation of the next response”, where the follow-up question or query is interpreted as second query as claimed)
and the answer comprises a summary of the subset of the chunks (para.[0078]; “answer may be a summary, a description of the results 176, or another text response that answers the prompt 170 based on the results”).
the summary comprising a non-verbatim summary of a paragraph of the document of the documents (para.[0097]; “result data set and asked to generate a summary. In this interaction, the AI/ML model receives the result data set that
generally includes aggregated or composite information specifically answering the user's question”).
and displaying, via the chatbot interface, a retrieved context regarding the document of the documents (para.[0090]; “interface of the chatbot can be provided together with a document viewer, for example, in a sidebar or tab shown concurrently with the document viewer interface. This arrangement can enable the user to view a document, such as a dashboard related to a dataset, while concurrently having a conversation with a chatbot designed to answer questions about the dataset”).
wherein: the retrieved context comprises information enabling a user to verify a source of the answer (para.[0032]; para.[0087]; “visualization data 190 can include the metadata for the chatbot response in which the visualization was generated”, para.[0105]; “The metadata 222 can also indicate information about the data source and data objects used to generate the chatbot response ….. the metadata 222 can specify
the source of data used to generate the chatbot response”, and para.[0116]; “retrieving the results, the metadata for a chatbot response”)
and the chatbot interface comprises a graphical user interface configured to display the retrieved context in a context retrieval drop-down menu (Fig.5. and para.[0090]; “interface of the chatbot can then be invoked by interacting with an icon or menu item for the chatbot”, and para.[0127]; “user interface 500 for a user to interact with the chatbot. A main area 510 includes various user prompts 511a-5llc and corresponding responses 512a-512c of the chatbot, along with a prompt entry field 514. In addition, the user interface 500 includes a snapshot area 520 with a search control 520 and several saved snapshots 522a-522c that the user previously saved”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate artificial intelligence chatbots of the system of Courcelle into classification of unstructured oilfield document of the system of Gupta to provide an interface for processing follow up query, thereby limiting the amount of information required to process the query.
As per claim 2, the rejection of claim 1 is incorporated and further Gupta et al (IN 202121000983 A) discloses,
wherein the documents comprise unstructured data, and wherein the unstructured data comprises text directed to the oil and gas exploration, drilling, or production (para.[0001]; “unstructured oilfield documents can include well completion reports, 5 daily drilling reports, production reports” and para.[0017]; “oilfield documents generated through the various exploration and production (E&P) operations of an oilfield”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate classification of unstructured oilfield document of the system of Gupta into generative AI pipelines that enable parallel large language model
(LLM) queries of the system of Mondlock to extract and arrange relevant unstructured oilfield data, thereby providing data that is easy to interpret by the user.
.
As per claim 3, the rejection of claim 2 is incorporated and further Gupta et al (IN 202121000983 A) and Mondlock et al (US 12,039,263 B1) discloses,
further comprising converting the documents from a first document format into a second document format (Gupta: para.[0058]; “OCR engine is executed on the document to convert the document to a document with recognized text”).
wherein the documents in the second document format are split into the chunks (Mondlock: col.6 lines 33 – 34; “splitting documents or assets into chunks”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate classification of unstructured oilfield document of the system of Gupta into generative AI pipelines that enable parallel large language model
(LLM) queries of the system of Mondlock to extract and arrange relevant unstructured oilfield data, thereby providing data that is easy to interpret by the user.
As per claim 4, the rejection of claim 3 is incorporated and further Gupta et al (IN 202121000983 A) discloses,
wherein converting the documents comprises performing optical character recognition (OCR) on the unstructured data in a portable document format (PDF) to convert the unstructured data into a text format (para.[0058]; “oilfield document is a scanned document, then an OCR engine is executed on the document to convert the document to a document with recognized text” and para.[0064]; “the table contents are extracted from the table …… Once rows and columns are identified, entries are determined from the rows and columns and transformed into a column separated file”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate classification of unstructured oilfield document of the system of Gupta into generative AI pipelines that enable parallel large language model
(LLM) queries of the system of Mondlock to extract and arrange relevant unstructured oilfield data, thereby providing data that is easy to interpret by the user.
As per claim 5, the rejection of claim 1 is incorporated and further Mondlock et al (US 12,039,263 B1) discloses,
wherein each embedding corresponds to a different one of the chunks, wherein the embeddings are generated using a deep learning model, and wherein the embeddings comprise multi-dimensional vectors in a form of real numbers (col.4 lines 65 – 67; “artificial intelligence (AI) algorithm that uses deep learning techniques to perform a number of natural language processing (NLP) tasks” and col.6 lines 50 – 52; “The embeddings represent the text chunks and data chunks as multi-dimensional ( e.g., 768 or 1,536 dimension) vectors of numerical values”).
As per claim 6, the rejection of claim 5 is incorporated and further Mondlock et al (US 12,039,263 B1) discloses,
wherein the query embedding is generated using the deep learning model (col.4 lines 65 – 67; “artificial intelligence (AI) algorithm that uses deep learning techniques to perform a number of natural language processing (NLP) tasks”).
As per claim 7, the rejection of claim 1 is incorporated and further Mondlock et al (US 12,039,263 B1) discloses,
wherein the subset of the chunks is retrieved using an approximate nearest neighbor algorithm (col.14 lines 39 – 41; “performing a k-nearest neighbors (KNN) search of the user query embedding”).
As per claim 8, the rejection of claim 1 is incorporated and further Mondlock et al (US 12,039,263 B1) discloses,
further comprising displaying the natural language query and the answer (col.6 lines 17 – 18; “GUI may contain an output text field for displaying the answer”).
As per claim 9, the rejection of claim 1 is incorporated and further Gupta et al (IN 202121000983 A) discloses,
further comprising performing a wellsite action in response to the answer (para.[0021]; “during early exploration stages, seismic data (161) may be gathered from the surface to identify possible locations of hydrocarbons” and para.[0024]; “the wellsite system components in response to data received, stored, processed, and/or analyzed, ….. to control and/or optimize various field operations”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate classification of unstructured oilfield document of the system of Gupta into generative AI pipelines that enable parallel large language model
(LLM) queries of the system of Mondlock to extract and arrange relevant unstructured oilfield data, thereby providing data that is easy to interpret by the user.
As per claim 10, the rejection of claim 9 is incorporated and further Gupta et al (IN 202121000983 A) discloses,
wherein the wellsite action comprises varying a weight or torque on a drill bit that is drilling a wellbore (para.[0021]; “during early exploration stages, seismic data (161) may be gathered from the surface to identify possible locations of hydrocarbons”, para.[0023]; “wellsite equipment configured to perform wellbore operations, such as logging, drilling, fracturing, production, or other applicable operations. ……. the wellsite system (192) is associated with a rig (132), a wellbore (112), and drilling equipment to perform drilling operation”, para.[0024]; “the wellsite system components in response to data received, stored, processed, and/or analyzed, ….. to control and/or optimize various field operations”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate classification of unstructured oilfield document of the system of Gupta into generative AI pipelines that enable parallel large language model
(LLM) queries of the system of Mondlock to extract and arrange relevant unstructured oilfield data, thereby providing data that is easy to interpret by the user.
As per claim 22, the rejection of claim 1 is incorporated and further Courcelle et al (US 2025/0293998 A1) discloses,
further comprising displaying the natural language query, wherein: the natural language query comprises a first text including "what is matter"; and the retrieved context in the context retrieval drop-down menu comprises a second text including "what is the matter?" (Fig.5. and para.[0090]; “interface of the chatbot can then be invoked by interacting with an icon or menu item for the chatbot”, and para.[0127]; “user interface 500 for a user to interact with the chatbot. A main area 510 includes various user prompts 511a-5llc and corresponding responses 512a-512c of the chatbot, along with a prompt entry field 514. In addition, the user interface 500 includes a snapshot area 520 with a search control 520 and several saved snapshots 522a-522c that the user previously saved”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate artificial intelligence chatbots of the system of Courcelle into classification of unstructured oilfield document of the system of Gupta to provide an interface for processing follow up query, thereby limiting the amount of information required to process the query.
8. Claims 11, 13 – 14, and 16 - 20 are rejected under 35 U.S.C. 103 as being unpatentable over Gupta et al (IN 202121000983 A), in view of Mondlock et al (US 12,039,263 B1), and further in view of Courcelle et al (US 2025/0293998 A1).
As per claim 11, Gupta et al (IN 202121000983 A) discloses,
A computing system, comprising: one or more processors; and a memory system comprising one or more non-transitory computer-readable media storing instructions that, when executed by at least one of the one or more processors, cause the computing system to perform operations (para.[0071]; “the computing system (600) may include one or more computer processors (602), non-persistent storage (604) (e.g., volatile memory, such as random access memory (RAM), cache memory)” and para.[0072]; “The computer processor(s) (602) may be an integrated circuit for processing instructions. For example, the computer processor(s) may be
one or more cores or micro-cores of a processor”).
the operations comprising: receiving documents (para.[0001]; “In the oil and gas industry, huge amount of data is received from the field in the form of digital and scanned unstructured oilfield documents”).
wherein the documents comprise unstructured data (para.[0001]; “the unstructured oilfield documents can include well completion reports, daily drilling reports, production reports”).
wherein the unstructured data comprises text directed to oil and gas exploration, drilling, or production (para.[0001]; “the unstructured oilfield documents can include well completion reports, daily drilling reports, production reports”).
converting the documents from a first document format into a second document format (para.[0058]; “OCR engine is executed on the document to convert the document to a document with recognized text”).
wherein converting the documents comprises performing optical character
recognition (OCR) on the unstructured data in a portable document format (PDF) to convert the unstructured data into a text format (para.[0058]; “oilfield document is a scanned document, then an OCR engine is executed on the document to convert the document to a document with recognized text” and para.[0064]; “the table contents are extracted from the table …… Once rows and columns are identified, entries are determined from the rows and columns and transformed into a column separated file”).
receiving a natural language query directed to the oil and gas exploration, drilling, or production (para.[0037]; “the document content type may be drilling log, production log, daily production report, well completion report, geochemistry lab report, seismic report, or other document content type”, para.[0086]; “extraction criteria may be as simple as an identifier string or may be a query”, and para.[0089]; ““user, or software application, may submit a statement or query into the DBMS ……… perform computations to respond to the query. The DBMS may return the result(s) to the user or software application”).
Gupta does not specifically disclose splitting the documents in the second document format into chunks, generating a plurality of embeddings based upon the chunks, wherein each embedding corresponds to a different one of the chunks, wherein the embeddings are generated using a deep learning model, and wherein the embeddings comprise multi-dimensional vectors in a form of real numbers, storing the chunks, the embeddings, and associated metadata in a vector database, generating a query embedding based upon the natural language query, wherein the query embedding is generated using the deep learning model, retrieving a subset of the chunks based upon the query embedding, wherein the subset of the chunks is retrieved using an approximate nearest neighbor algorithm, and generating an answer in response to the natural language query, wherein the answer is based upon the natural language query and the subset of the chunks.
However, Mondlock et al (US 12,039,263 B1) in an analogous art discloses,
splitting the documents in the second document format into chunks (col.6 lines 33 – 34; “splitting documents or assets into chunks”).
generating embeddings based upon the chunks (col.6 lines 33 – 34; “splitting documents or assets into chunks and generating embeddings of those chunks” and col.6 lines 49 – 50; “generating embeddings from each text chunk and/or data chunk”).
wherein each embedding corresponds to a different one of the chunks (col.6 lines 37 – 41; “split each asset of asset collections 146 and asset collections 164 into a plurality of text chunks and/or data chunks. The text chunks 40 may be paragraph-sized, sentence-sized, fixed-sized ( e.g., 50 words) or any other appropriate size”).
wherein the embeddings are generated using a deep learning model (col.4 lines 65 – 66; “artificial intelligence (AI) algorithm that uses deep learning techniques”).
and wherein the embeddings comprise multi-dimensional vectors in a form of real numbers (col.6 lines – 52; “The embeddings represent the text chunks and data chunks as multi-dimensional ( e.g., 768 or 1,536 dimension) vectors of numerical values”).
storing the chunks, the embeddings, and associated metadata in a vector database (col.6 lines 60 – 63; “document/asset/expert module 122 may save the embeddings into embeddings 168 in the external data sources 160. The embeddings 168 may comprise a vector database”).
receiving a natural language query directed to the oil and gas exploration, drilling, or production (col.7 lines 10 – 12; “receive the query or the query plus retrieved chat history and classify the query into one or more of a plurality of pre-defined intents”).
generating a query embedding based upon the natural language query (col.7 lines 15 – 18; “generating an embedding of each predefined intent; (2) generating an embedding of the user query”).
wherein the query embedding is generated using the deep learning model (col.4 lines 65 – 66; “artificial intelligence (AI) algorithm that uses deep learning techniques”).
retrieving a subset of the chunks based upon the query embedding (col.8 lines 13 – 15; “generating an embedding of the user query; and (2) compare the user query embedding to the document and asset embeddings in embeddings” and col.8 lines 41 – 44; “compare the embedding of the user query to embeddings of the text chunks and/or data chunks to identify relevant text chunks and/or relevant data chunks”).
wherein the subset of the chunks is retrieved using an approximate nearest neighbor algorithm (col.14 lines 39 – 41; “performing a k-nearest neighbors (KNN) search of the user query embedding”).
and generating an answer in response to the natural language query (col.8 lines 17 – 21; “relevant information identification module 130 may transmit the query embedding, the document embeddings, and/or the data 20 embeddings, via the LLM interface module 132, to the LLM Service” and col.8 lines 25 – 32; “topic modeling may include (1) performing topic modeling on the user query to identify one or more topic keywords; and (2) searching the document collections 144, asset collections 146, expert collections 148, document collections 162, asset collections 164, and/or expert collections 166 with the topic keywords to identify relevant documents, assets, and/or experts”).
wherein the answer is based upon the natural language query and the subset of the chunks (col.8 lines 50 – 57; “the LLM interface module 132 may transmit prompts to and receive answers from the LLM service ………… receive relevant information from the relevant text chunk and/or relevant data chunk” and col.16 lines 19 – 21; “prompts may cause the LLM service 170 to output relevant information for each of
the text chunks and data chunks”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate generative AI pipelines that enable parallel large language model (LLM) queries of the system of Mondlock into classification of unstructured oilfield document of the system of Gupta to supplements user queries with relevant supplied data to enable LLMs to provide improved responses that tailor towards efficient analysis of oilfield data.
Neither Gupta nor Mondlock does not specifically disclose the generating the answer comprises generating a chatbot interface; the query is a first query; the operations further comprise receiving, via the chatbot interface, a second query regarding a document of the documents; and the answer comprises a summary of the subset of the chunks, the summary comprising a non-verbatim summary of a paragraph of the document of the documents; and displaying, via the chatbot interface, a retrieved context regarding the document of the documents, wherein the retrieved context comprises information verifying a source of the summary of the subset of the chunks..
However, Courcelle et al (US 2025/0293998 A1) in an analogous art discloses,
the generating the answer comprises generating a chatbot interface (para.[0084]; “generates and sends a response 182 as the answer of the chatbot to the user prompt 170. The response 182 is then displayed on the user interface”).
the query is a first query; the operations further comprise receiving, via the chatbot interface (para.[0061]; “user 105c interacts with the chatbot by submitting the user prompt 170, such as a natural language question or statement” and para.[0093]; “a user prompt such as a natural language query is received”).
a second query regarding a document of the documents (para.[0022]; “ask follow-up questions about that conversation result” and para.[0028]; “the user can immediately ask a follow-up question, and the chatbot can use the information from the previous response (including potential intermediate information or information derived from multiple prior interactions) to inform the creation of the next response”, where the follow-up question or query is interpreted as second query as claimed);
and the answer comprises a summary of the subset of the chunks (para.[0078]; “answer may be a summary, a description of the results 176, or another text response that answers the prompt 170 based on the results”).
the summary comprising a non-verbatim summary of a paragraph of the document of the documents (para.[0097]; “result data set and asked to generate a summary. In this interaction, the AI/ML model receives the result data set that generally includes aggregated or composite information specifically answering the user's question”).
and displaying, via the chatbot interface, a retrieved context regarding the document of the documents (para.[0090]; “interface of the chatbot can be provided together with a document viewer, for example, in a sidebar or tab shown concurrently with the document viewer interface. This arrangement can enable the user to view a document, such as a dashboard related to a dataset, while concurrently having a conversation with a chatbot designed to answer questions about the dataset”).,
wherein the retrieved context comprises information verifying a source of the summary of the subset of the chunks (fig.5. and para.[0090]; “interface of the chatbot can then be invoked by interacting with an icon or menu item for the chatbot”, and para.[0127]; “user interface 500 for a user to interact with the chatbot. A main area 510 includes various user prompts 511a-5llc and corresponding responses 512a-512c of the chatbot, along with a prompt entry field 514. In addition, the user interface 500 includes a snapshot area 520 with a search control 520 and several saved snapshots 522a-522c that the user previously saved”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate artificial intelligence chatbots of the system of Courcelle into classification of unstructured oilfield document of the system of Gupta to provide an interface for processing follow up query, thereby limiting the amount of information required to process the query.
As per claim 13, the rejection of claim 11 is incorporated and further Gupta et al (IN 202121000983 A) discloses,
wherein the answer is generated by a large language model (LLM), wherein the LLM has access to domain-specific documents that comprise the text directed to the oil and gas exploration, drilling, or production, and wherein the LLM is not trained using the domain-specific documents (para.[0001]; “unstructured oilfield documents can include well completion reports, 5 daily drilling reports, production reports” and para.[0017]; “oilfield documents generated through the various exploration and production (E&P) operations of an oilfield”).
As per claim 14, the rejection of claim 11 is incorporated and further Mondlock et al (US 12,039,263 B1) discloses,
wherein the answer is also based upon a system prompt, and wherein the system prompt comprises instructions for how to answer the natural language query (col.1 lines 34 – 37; “receiving a user query, fetching relevant external data, and submitting prompts to cause the LLM to answer the user query based on provided relevant external data”, where external data is interpreted as “oil and gas exploration drilling, and/or production” as claimed and col.8 lines 50 – 57; “the LLM interface module 132 may transmit prompts to and receive answers from the LLM service”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate generative AI pipelines that enable parallel large language model (LLM) queries of the system of Mondlock into classification of unstructured oilfield document of the system of Gupta to supplements user queries with relevant supplied data to enable LLMs to provide improved responses that tailor towards efficient analysis of oilfield data.
As per claim 16, Gupta et al (IN 202121000983 A) discloses,
A non-transitory computer-readable medium storing instructions that, when executed by one or more processors of a computing system, cause the computing system to perform operations (para.[0075]; “computer readable program code to perform embodiments of the disclosure may be stored, in whole or in part, temporarily or permanently, on a non-transitory computer readable medium …… computer readable program code that, when executed by a processor(s), is configured to perform one or more embodiments of the disclosure”).
the operations comprising: receiving documents (para.[0001]; “In the oil and gas industry, huge amount of data is received from the field in the form of digital and scanned unstructured oilfield documents”).
wherein the documents comprise unstructured data (para.[0001]; “the unstructured oilfield documents can include well completion reports, daily drilling reports, production reports”).,
wherein the unstructured data comprises text directed to oil and gas exploration, drilling, or production (para.[0001]; “the unstructured oilfield documents can include well completion reports, daily drilling reports, production reports”).
converting the documents from a first document format into a second document format (para.[0058]; “OCR engine is executed on the document to convert the document to a document with recognized text”).
wherein converting the documents comprises performing optical character recognition (OCR) on the unstructured data in a portable document format (PDF) to convert the unstructured data into a text format (para.[0058]; “oilfield document is a scanned document, then an OCR engine is executed on the document to convert the document to a document with recognized text” and para.[0064]; “the table contents are extracted from the table …… Once rows and columns are identified, entries are determined from the rows and columns and transformed into a column separated file”).
receiving a natural language query directed to oil and gas exploration, drilling, or production (para.[0016]; “extract table information from one or more tables in the oilfield document” and para.[0086]; “extraction criteria may be as simple as an identifier string or may be a query”).
Gupta does not specifically disclose splitting the unstructured data in the second document format into chunks; generating a plurality of embeddings based upon the chunks, wherein each embedding corresponds to a different one of the chunks, wherein the embeddings are generated using a deep learning model, and wherein the embeddings comprise multi-dimensional vectors in a form of real numbers; storing the chunks, the embeddings, and associated metadata in a vector database; receiving a natural language query directed to oil and gas exploration, drilling, or production; generating a query embedding based upon the natural language query, wherein the query embedding is generated using the deep learning model; retrieving a subset of the chunks based upon the query embedding, wherein the subset of the chunks is retrieved using an approximate nearest neighbor algorithm; and generating an answer in response to the natural language query, wherein the answer is based upon the natural language query, the subset of the chunks, and a system prompt, the answer comprises the subset of the chunks and a summary of the subset of the chunks, the answer is generated by a large language model (LLM), the LLM has access to domain-specific documents that comprise text directed to oil and gas exploration, drilling, or production, the LLM is not trained using the domain-specific documents, and the system prompt comprises instructions for how to answer the natural language query.
However, Mondlock et al (US 12,039,263 B1) in an analogous art discloses,
splitting the unstructured data in the second document format into chunks (col.6 lines 33 – 34; “splitting documents or assets into chunks”).
generating embeddings based upon the chunks (col.6 lines 33 – 34; “splitting documents or assets into chunks and generating embeddings of those chunks” and col.6 lines 49 – 50; “generating embeddings from each text chunk and/or data chunk”).
wherein each embedding corresponds to a different one of the chunks (col.6 lines 37 – 41; “split each asset of asset collections 146 and asset collections 164 into a plurality of text chunks and/or data chunks. The text chunks 40 may be paragraph-sized, sentence-sized, fixed-sized ( e.g., 50 words) or any other appropriate size”).
wherein the embeddings are generated using a deep learning model (col.4 lines 65 – 66; “artificial intelligence (AI) algorithm that uses deep learning techniques”).
and wherein the embeddings comprise multi-dimensional vectors in a form of real numbers (col.6 lines – 52; “The embeddings represent the text chunks and data chunks as multi-dimensional ( e.g., 768 or 1,536 dimension) vectors of numerical values”).
storing the chunks, the embeddings, and associated metadata in a vector database (col.6 lines 60 – 63; “document/asset/expert module 122 may save the embeddings into embeddings 168 in the external data sources 160. The embeddings 168 may comprise a vector database”).
receiving a natural language query directed to the oil and gas exploration, drilling, or production (col.7 lines 10 – 12; “receive the query or the query plus retrieved chat history and classify the query into one or more of a plurality of pre-defined intents”).
generating a query embedding based upon the natural language query (col.7 lines 15 – 18; “generating an embedding of each predefined intent; (2) generating an embedding of the user query”).
wherein the query embedding is generated using the deep learning model (col.4 lines 65 – 66; “artificial intelligence (AI) algorithm that uses deep learning techniques”).
retrieving a subset of the chunks based upon the query embedding (col.8 lines 13 – 15; “generating an embedding of the user query; and (2) compare the user query embedding to the document and asset embeddings in embeddings” and col.8 lines 41 – 44; “compare the embedding of the user query to embeddings of the text chunks and/or data chunks to identify relevant text chunks and/or relevant data chunks”).
wherein the subset of the chunks is retrieved using an approximate nearest neighbor algorithm (col.14 lines 39 – 41; “performing a k-nearest neighbors (KNN) search of the user query embedding”).
and generating an answer in response to the natural language query (col.8 lines 17 – 21; “relevant information identification module 130 may transmit the query embedding, the document embeddings, and/or the data 20 embeddings, via the LLM interface module 132, to the LLM Service” and col.8 lines 25 – 32; “topic modeling may include (1) performing topic modeling on the user query to identify one or more topic keywords; and (2) searching the document collections 144, asset collections 146, expert collections 148, document collections 162, asset collections 164, and/or expert collections 166 with the topic keywords to identify relevant documents, assets, and/or experts”).
wherein the answer is based upon the natural language query, the subset of the chunks and a system prompt (col.8 lines 50 – 57; “the LLM interface module 132 may transmit prompts to and receive answers from the LLM service ………… receive relevant information from the relevant text chunk and/or relevant data chunk” and col.16 lines 19 – 21; “prompts may cause the LLM service 170 to output relevant information for each of the text chunks and data chunks”).
the answer comprises the subset of the chunks and a summary of the subset of the chunks (col.7 lines 30 – 32; “The intents may be used to select which document, asset, or expert sources will be used to answer the query. ……… the user query may ask for a financial summary”).
the answer is generated by a large language model (LLM) (col.8 lines 50 – 57; “the LLM interface module 132 may transmit prompts to and receive answers from the LLM service”).
the LLM has access to domain-specific documents that comprise the text directed to the oil and gas exploration, drilling, or production (col.1 lines 34 – 37; “receiving a user query, fetching relevant external data, and submitting prompts to cause the LLM to answer the user query based on provided relevant external data” and col.13 lines 50 -52; “retrieving the one or more expert biographies from the one or more selected expert collections. The expert biographies may be retrieved by the document/asset/expert”).
the LLM is not trained using the domain-specific documents and the system prompt comprises instructions for how to answer the natural language query (col.1 lines 34 – 37; “receiving a user query, fetching relevant external data, and submitting prompts to cause the LLM to answer the user query based on provided relevant external data”, where external data is interpreted as “oil and gas exploration drilling, and/or production” as claimed, col.8 lines 3 – 5; “text instructing the LLM service 170 to answer the user query with the supplemental information instead of relying”, and col.8 lines 50 – 57; “the LLM interface module 132 may transmit prompts to and receive answers from the LLM service”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate generative AI pipelines that enable parallel large language model (LLM) queries of the system of Mondlock into classification of unstructured oilfield document of the system of Gupta to supplements user queries with relevant supplied data to enable LLMs to provide improved responses that tailor towards efficient analysis of oilfield data.
Neither Gupta nor Mondlock does not specifically disclose the generating the answer comprises generating a chatbot interface; the query is a first query; the operations further comprise receiving, via the chatbot interface, a second query regarding a document of the documents; and the summary of the subset of the chunks comprises a non-verbatim summary of a paragraph of the document of the documents, displaying, via the chatbot interface, a retrieved context regarding the document of the documents, wherein the retrieved context comprises information enabling a user to verify a source of the answer; and displaying, via the chatbot interface, a graphical user interface showing the retrieved context in a context retrieval drop-down menu.
However, Courcelle et al (US 2025/0293998 A1) in an analogous art discloses,
the generating the answer comprises generating a chatbot interface (para.[0084]; “generates and sends a response 182 as the answer of the chatbot to the user prompt 170. The response 182 is then displayed on the user interface”).
the query is a first query; the operations further comprise receiving, via the chatbot interface (para.[0061]; “user 105c interacts with the chatbot by submitting the user prompt 170, such as a natural language question or statement” and para.[0093]; “a user prompt such as a natural language query is received”).
a second query regarding a document of the documents (para.[0022]; “ask follow-up questions about that conversation result” and para.[0028]; “the user can immediately ask a follow-up question, and the chatbot can use the information from the previous response (including potential intermediate information or information derived from multiple prior interactions) to inform the creation of the next response”, where the follow-up question or query is interpreted as second query as claimed).
and the summary of the subset of the chunks comprises a non-verbatim summary of a paragraph of the document of the documents (para.[0097]; “result data set and asked to generate a summary. In this interaction, the AI/ML model receives the result data set that generally includes aggregated or composite information specifically answering the user's question”).
,
displaying, via the chatbot interface, a retrieved context regarding the document of the documents (para.[0090]; “interface of the chatbot can be provided together with a document viewer, for example, in a sidebar or tab shown concurrently with the document viewer interface. This arrangement can enable the user to view a document, such as a dashboard related to a dataset, while concurrently having a conversation with a chatbot designed to answer questions about the dataset”).
wherein the retrieved context comprises information enabling a user to verify a source of the answer (para.[0032]; para.[0087]; “visualization data 190 can include the metadata for the chatbot response in which the visualization was generated”, para.[0105]; “The metadata 222 can also indicate information about the data source and data objects used to generate the chatbot response ….. the metadata 222 can specify
the source of data used to generate the chatbot response”, and para.[0116]; “retrieving the results, the metadata for a chatbot response”).
and displaying, via the chatbot interface, a graphical user interface showing the retrieved context in a context retrieval drop-down menu (fig.5. and para.[0090]; “interface of the chatbot can then be invoked by interacting with an icon or menu item for the chatbot”, and para.[0127]; “user interface 500 for a user to interact with the chatbot. A main area 510 includes various user prompts 511a-5llc and corresponding responses 512a-512c of the chatbot, along with a prompt entry field 514. In addition, the user interface 500 includes a snapshot area 520 with a search control 520 and several saved snapshots 522a-522c that the user previously saved”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate artificial intelligence chatbots of the system of Courcelle into classification of unstructured oilfield document of the system of Gupta to provide an interface for processing follow up query, thereby limiting the amount of information required to process the query.
As per claim 17, the rejection of claim 16 is incorporated and further Mondlock et al (US 12,039,263 B1) discloses,
wherein the system prompt is optimized to provide accurate answers on a dedicated subject matter assessment (col.12 lines 63 – 64; “user query and intent may include prompt engineering to optimize the final query”).
As per claim 18, the rejection of claim 16 is incorporated and further Mondlock et al (US 12,039,263 B1) discloses,
wherein the system prompt is optimized by making iterative improvements and programmatic improvements (col.23 lines 55 – 59; “The augmented user query may include each of the relevant information responses, the user query, and a prompt causing the LLM to generate an answer. The prompt may cause the LLM to generate the answer”).
As per claim 19, the rejection of claim 16 is incorporated and further Mondlock et al (US 12,039,263 B1) discloses,
wherein the answer is 50 words or less and contains names of commercial products relevant for a specific scenario identified in the natural language query (col.7 lines 32 – 35; “the user query may ask for a financial summary of Acme Corp., and the intent classification module 126 may classify the intent of the user query as public company financial information”).
As per claim 20, the rejection of claim 16 is incorporated and further Gupta et al (IN 202121000983 A) discloses,
wherein the operations further comprise performing a wellsite action in response to the answer, wherein the wellsite action comprises generating and transmitting a signal that recommends, instructs, or causes a physical action to occur (para.[0021]; “during early exploration stages, seismic data (161) may be gathered from the surface to identify possible locations of hydrocarbons”, para.[0023]; “wellsite equipment configured to perform wellbore operations, such as logging, drilling, fracturing, production, or other applicable operations. ……. the wellsite system (192) is associated with a rig (132), a wellbore (112), and drilling equipment to perform drilling operation”, para.[0024]; “the wellsite system components in response to data received, stored, processed, and/or analyzed, ….. to control and/or optimize various field operations”).
Therefore, it would have been obvious to one of ordinary skill in the art before the invention was filed to incorporate classification of unstructured oilfield document of the system of Gupta into generative AI pipelines that enable parallel large language model
(LLM) queries of the system of Mondlock to extract and arrange relevant unstructured oilfield data, thereby proving data that is easier to interpret by the user.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
TITLE: Electronic data parsing and interactive user interfaces for data processing,
US 2018/0239959 A1 authors: Bui et al.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AUGUSTINE KUNLE OBISESAN whose telephone number is (571)272-2020. The examiner can normally be reached 9:00am - 5:00.
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, Ajay Bhatia can be reached at (571) 272-3906. 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.
/AUGUSTINE K. OBISESAN/
Primary Examiner
Art Unit 2156
9/28/2026
Prosecution Insights