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 .
Status of Claims
This is first office action on the merits in response to the response to restriction requirements filed on 05/08/2026.
Claim 1-18 are currently pending. Group I comprising claims 1-7 have been elected. Claims 8-18 are directed to the non-elected Groups II and III. Claim 1-7 have been examined.
Priority
Applicant's claim for the benefit of a US Patent Provisional Application No. 63/511,465 filed on 06/30/2023 is acknowledged. Applicant's claim for the benefit of a US Patent Application No. 18/305,969 filed on 04/24/2023 is acknowledged. Applicant's claim for the benefit of a US Patent Provisional Application No. 63/334,527 filed on 04/25/2022 is acknowledged. Applicant's claim for the benefit of a United States Patent Application No. 16/600,376 filed on 10/11/2019 is acknowledged. Applicant's claim for the benefit of a United States Patent Application No. 16/555,611 filed on 08/29/2019 is acknowledged. Applicant's claim for the benefit of a US Patent Provisional Application No. 62/829,151 filed on 04/04/2019 is acknowledged. Applicant's claim for the benefit of a US Patent Provisional Application No. 62/769,024 filed on 11/19/2018 is acknowledged.
The effective filing date of claims 1-7 of the instant application is 06/30/2023 because no prior-filed application discusses the inventive concept of “dynamically redefining a domain of the function to include dark data.”
Claim Rejections - 35 USC § 102(a)(2)
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1-5 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Statton (US 20240370339).
Regarding Claim 1, Statton teaches a artificial intelligence implemented method for executing a function by dynamically redefining a domain of the function to include dark data stored in a database, wherein the artificial intelligence instantiates data events in an in-memory neural network in accordance with the domain so as to execute the function, wherein the database includes a set of data events stored therein, wherein each data event is defined by a one or more values, wherein each value is associated with a category of a set of possible categories, wherein the function is mapped to a predefined set of relevant data events and a predefined set of relevant categories so as to initially define the domain of the function, wherein the set of relevant categories is a subset of the set of possible categories, wherein the set of relevant data events is a subset of the set of data events in which at least one data event has at least one relevant value, wherein a relevant value is a value associated with relevant category (Paragraphs 0098, 0064, and 0106 teach FIG. 4 is a flow diagram illustrating an example operation of a computing system, in accordance with one or more techniques of this disclosure; FIG. 3 is a block diagram illustrating an example of response generation platform and operations in further detail; response generation platform includes interface module, filter generator, embeddings generator, and query processor; interface module may execute an interface by which other systems or devices may communicate with response generation platform; the interface presented by interface module may be a gRPC, HTTP, RESTful, command-line, graphical user, web, or other interface; FIG. 5 depicts a process of filter provision/generation and embeddings generation), the artificial intelligence implemented method comprising: identifying one or more discrete values from among the relevant values of the relevant data events (Paragraphs 0098, 0065-0067, and 0106 teach a computing system initially may process an input to generate (or otherwise obtain) a filter, wherein the input indicates a context for one or more queries; interface receives input that includes text or other data that indicates context for one or more queries from a user or application; the one or more queries may include input; filter generator processes input to determine types of data relevant to the queries; for example, filter generator may analyze the input using a machine learning model to decode the types of data the user is interested in; types of data may be according to file type (e.g., Email data, File Share data, Databases, or other unstructured data), according to association with certain entities (persons, organizations, etc.), according to time or dates, according to topic or semantic similarly, according to context, or other dimensions in which to categorize or characterize data, such as text data; in this example, filter generator applies a machine learning model to analyze input to decode the types of data relevant to the queries; this analysis allows the system to accurately determine the user's intent and tailor the filter accordingly, ensuring that the most relevant data is processed and included in the index of embeddings; machine learning model may include natural language processing (NLP) and deep learning algorithms, for instance; filter generator generates a filter based on the decoded data types; filter is designed to enable response generation platform to efficiently sift through large quantities of backup data in backups, focusing on the specific data types identified by the machine learning analysis applied by filter generator, and ensuring that only relevant data is processed and included in the index of embeddings; a user/administrator or application (via API access layer) by sending filter selection can use response generation platform to provide or generate (by dynamic filter service) a filter which can be applied to a dataset stored to storage system); identifying, in other data events, values that correspond to the discrete values, wherein the other data events are within the set of data events but are not relevant data events (Paragraphs 0098, 0068, and 0106 teach next, computing system may apply the filter to backup data to obtain filtered data from the backup data; response generator platform applies filter to backup data of backups to obtain filtered data that is likely relevant to the one or more queries, as indicated by input and determined by filter generator; in some cases, input may specify a particular backup of backups as context for the one or more queries (i.e., all filtered data is included in the specified backup); filter is interpreted by dynamic filter service and sent to data access layer to retrieve data satisfying filter from the dataset; the dataset may be one of backups, for instance, or another dataset); linking each discrete value to each category of the set of possible categories in which the associated value for any data event corresponds to the discrete value, so as to identify one or more dark categories as categories that are linked to one or more of the discrete values and are not in the set of relevant categories (Paragraphs 0098, 0069, 0072, and 0106 teach next, computing system may generate an index of embeddings from the filtered data; the Naive Bayes Classifier is a popular probabilistic algorithm for text classification problems; it is based on Bayes' theorem and assumes independence between the features; in the context of the filtering application, the algorithm calculates the probability of different categories or labels for the given text data; it assigns the class with the highest probability to the text, which can be seen as a filtering decision; where the class/label matches the filter generated by filter generator for input, the classified text data is relevant according to input; embeddings generator processes the obtained, filtered text data that matches the generated filter to generate index of embeddings; an embedding is a numerical-typically a vector-representation of a piece of information, for example, text, documents, images, audio, etc.; an embedding is a way of representing data as points in n-dimensional space so that similar data points cluster together; in NLP and other forms of artificial intelligence contexts, an embedding can represent text data and be used in text analysis because the embedding is dense in semantic meaning; once filtered by application of filter and retrieved by data access layer, embeddings generator applying model processes the filtered dataset through a vectorization engine to catalog the filtered dataset and store the resultant vectors into a vector database with subsequent metadata); mapping each dark category to the function, so as to redefine the domain (Paragraphs 0098, 0072, and 0106-0107 teach next, computing system may process, based on the index of embeddings, a query to generate a response for the query; embeddings generator obtains items of filtered text data (e.g., files, emails, text objects, etc.), encodes the items as embeddings, and indexes them to generate index of embeddings; based on index of embeddings, query processor processes the query to generate a response; for example, query processor can use index of embeddings to, e.g., perform semantic search of queries against index of embeddings and generate responses for queries based on the semantic search results; index of embeddings may adhere to a RBAC model allowing for access control over read, write, update, and deletion of the index at a role-level; this vector database of vectors is depicted in FIG. 5 and described elsewhere in this document as index of embeddings having embeddings; an entry into the vector database can either contain the full file, part of a file, or location of the file along with the embedding itself; additional metadata may be added, such as file location (for citation purposes supporting a response to a query) and Access Control List (ACL) information; database layer may either save the data/text chunks into embeddings or save a reference link to the data for retrieval); instantiating in the in-memory neural network data events according to the redefined domain (Paragraphs 0075, 0077, and 0094 teach in a further aspect of the invention, index of embeddings is made available to drive RAG queries, and other such AI/ML application usage from the user of application; RAG is a framework that combines pre-trained sequence-to-sequence (seq2seq) models with a dense retrieval mechanism, allowing for the generation of more informed and contextually relevant output; data platform may provide robust and domain-specific context to RAG-driven AI systems; by leveraging the robust file system, data platform incorporating response generation platform incorporates (or enables) ‘AI Ready’ for RAG-assisted large language models (LLMs) through an on-demand index of embeddings that are provided just-in-time to the application requesting the data; the data may be secured through RBAC control models; the system is designed to handle large quantities of backup data, making it suitable for use in a wide range of advanced filesystems and artificial intelligence applications); and executing the function in accordance with the redefined domain (Paragraph 0075, 0092, and 0106 teach this allows users and applications to retrieve data in a secure and efficient manner, without compromising the integrity of the system or the data itself; the RAG queries are also tailored to the specific data types identified by the machine learning analysis, ensuring that users and applications can quickly and easily access the desired information; by generating a unique filter and creating an index of embeddings 164 on-demand or “on the fly,” the system can quickly and efficiently process large quantities of backup data to be made available for RAG queries or other AI/ML applications, ensuring that users can access the information they need without significant delays; index of embeddings that is created can now function and interact with post-processing interactions such as answering user questions (“queries”); in some examples, data access layer is configured to redact personally identifiable information from data retrieved from storage system).
Regarding Claim 2, Statton teaches all the limitations of claim 1 above; and Statton further teaches wherein each discrete value is logically distinct from each other discrete value, and where each discrete value corresponds to logically equivalent relevant values (Paragraphs 0067-0069 teach filter generator generates a filter based on the decoded data types; filter is designed to enable response generation platform to efficiently sift through large quantities of backup data in backups, focusing on the specific data types identified by the machine learning analysis applied by filter generator, and ensuring that only relevant data is processed and included in the index of embeddings; response generator platform applies filter to backup data of backups to obtain filtered data that is likely relevant to the one or more queries, as indicated by input and determined by filter generator; input may specify a particular backup of backups as context for the one or more queries (i.e., all filtered data is included in the specified backup); first, the index of data for the data backup system is preprocessed; the preprocessing step helps to clean the backup data that is available to be filtered and processed, remove unnecessary noise, and make it suitable for further analysis; the preprocessing includes tokenization, stop word removal, and other text normalization techniques).
Regarding Claim 3, Statton teaches all the limitations of claim 1 above; and Statton further teaches wherein identifying the discrete values includes: associating each discrete value to corresponding relevant values via a first referential table (Paragraphs 0057 and 0059 teach backup manager may generate and manage chunk metadata for generating, viewing, retrieving, or restoring objects stored as chunks (and references thereto) within chunkfiles, for any of backups; stored objects may be represented and manipulated using logical files for identifying chunks for the objects; chunk metadata may include a chunk table that describes chunks ;the chunk table may include respective chunk IDs for chunks and may contain pointers to chunkfiles and offsets within chunkfiles for retrieving chunks from storage system; chunks are written into chunkfiles at different offsets; by comparing new chunk IDs to the chunk table, backup manager can determine if the data already exists on the system; if the chunks already exist, data can be discarded and metadata for an object updated to reference the existing chunk; backup manager may use the chunk table to look up the chunkfile identifier for the chunkfile that contains a chunk; chunk metadata may include a chunkfile table that describes respective physical or virtual locations of chunkfiles on storage system, along with other metadata about the chunkfile, such as a checksum, encryption data, compression data, etc.; in FIG. 2, backup manager causes backup metadata and chunk metadata to be stored to storage system; backup manager causes some or all of backup metadata and chunk metadata to be stored to storage system; backup manager, optionally in conjunction with file system manager, may use backup metadata, chunk metadata, and/or file system metadata to restore any of backups to a file system implemented by data platform, which may be presented by file system manager to other systems).
Regarding Claim 4, Statton teaches all the limitations of claim 1 above; and Statton further teaches wherein identifying, in other data events, values that correspond to the discrete values includes: associating the discrete values to corresponding values of the other data events via a second referential table (Paragraphs 0069-0070 and 0106 teach CountVectorizer is a technique used in natural language processing to transform text data into a matrix of token counts; it effectively creates a bag-of-words representation of the text data, where the occurrence of each word in the dataset is tracked; this structured representation of the text data is then used as input for the machine learning model; the Naive Bayes Classifier is a popular probabilistic algorithm for text classification problems; in the context of the filtering application, the algorithm calculates the probability of different categories or labels for the given text data; it assigns the class with the highest probability to the text, which can be seen as a filtering decision; where the class/label matches the filter generated by filter generator for input, the classified text data is relevant according to input; in summary, the filtering process in this example starts by preprocessing the text data and creating a structured representation using the CountVectorizer; this representation is used as input for the Naive Bayes Classifier, which then calculates the probability of different categories for the text data; the class with the highest probability is chosen as the final label, effectively filtering the text data from backups based on its content; FIG. 5 depicts a process of filter provision/generation and embeddings generation; a user/administrator or application (via API access layer) by sending filter selection can use response generation platform to provide or generate (by dynamic filter service) a filter which can be applied to a dataset stored to storage system; filter is interpreted by dynamic filter service and sent to data access layer to retrieve data satisfying filter from the dataset; the dataset may be one of backups, for instance, or another dataset; once filtered by application of filter and retrieved by data access layer, embeddings generator applying model processes the filtered dataset through a vectorization engine to catalog the filtered dataset and store the resultant vectors into a vector database with subsequent metadata; this vector database of vectors is depicted in FIG. 5 and described elsewhere in this document as index of embeddings having embeddings; an entry into the vector database can either contain the full file, part of a file, or location of the file along with the embedding itself; additional metadata may be added, such as file location (for citation purposes supporting a response to a query) and Access Control List (ACL) information; index of embeddings that is created can now function and interact with post-processing interactions such as answering user questions (“queries”); data access layer is configured to redact personally identifiable information from data retrieved from storage system).
Regarding Claim 5, Statton teaches all the limitations of claim 1 above; and Statton further teaches identifying dark-data events as the other data events having values that correspond to the discrete values (Paragraphs 0098 and 0101 teach FIG. 4 is a flow diagram illustrating an example operation of a computing system, in accordance with one or more techniques of this disclosure; computing system initially may process an input to generate (or otherwise obtain) a filter, wherein the input indicates a context for one or more queries; next, computing system may apply the filter to backup data to obtain filtered data from the backup data; data access layer is configured to receive an input and return data from storage system based on the dataset/filter that was requested as the input to data access layer); and identifying dark-data event values as the values of the dark-data events (Paragraphs 0098, 0100, and 0102 teach next, computing system may generate an index of embeddings from the filtered data; filter may be generated by either an end-user or an automated system, e.g., dynamic filter service; dynamic filter service may be an example instance of filter generator; filtered data is retrieved from data access layer, e.g., as a view or snapshot of file system or filtered portion thereof at a particular time; the filtered data is presented to database layer; dynamic filter service is configured to receive an input and convert it into a filter (e.g., filter) when requesting data from data access layer).
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 6-7 are rejected under 35 U.S.C. 103 as being unpatentable over Statton (US 20240370339) in view of Tibbet (US 20210081477).
Regarding Claim 6, Statton teaches all the limitations of claim 1 above; however, Statton does not explicitly teach wherein the function includes detecting causes of outlier data-event scenarios, and wherein the relevant data-events are data-events corresponding to predefined outlier data-event scenarios.
Tibbet from same or similar field of endeavors teaches wherein the function includes detecting causes of outlier data-event scenarios, and wherein the relevant data-events are data-events corresponding to predefined outlier data-event scenarios (Paragraph 0302 teaches some or all of the prior event information can be used to determine whether a current event is an anomaly; an event is classified as “close to” (or potentially) an anomaly; events that are “close to” an anomaly, can be stored and reviewed periodically to determine if classification as an anomaly is appropriate based on the sufficiency of further information).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified Statton to incorporate the teachings of Tibbet for the function to include detecting causes of outlier data-event scenarios, and wherein the relevant data-events are data-events corresponding to predefined outlier data-event scenarios.
There is motivation to combine Tibbet into Statton because classifying an event as a “near anomaly” may be appropriate when there the initial number of signals is insufficient and there is not enough information to make a reliable determination. “Near anomalies” are also useful for review to update models for future use (Tibbet Paragraph 0296).
Regarding Claim 7, the combination of Statton and Tibbet teaches all the limitations of claim 6 above; however, the combination does not explicitly teach wherein executing the function in accordance with the redefined domain comprises: applying one or more rules to the redefined domain so as to determine membership in a truth category; generating a computed class based on the truth category membership, wherein the computed class associates, for each relevant data-event member of the truth category, the values of the data-event for each category of the computed class; neutrosophically analyzing the computed class using multi-level regression analysis to identify (a) one or more first level outlier data-event scenarios for each category of the computed class, and (b) one or more next level outlier data-event scenarios for each category of the computed class, so as to determine a plurality of unique multi-level outlier data-event scenarios; and neutrosophically analyzing the plurality of unique multi-level outlier data-event scenarios to identify one or more systemic occurrences of data-event scenarios that do not correspond to the predefined outlier data-event scenarios.
Tibbet further teaches wherein executing the function in accordance with the redefined domain comprises: applying one or more rules to the redefined domain so as to determine membership in a truth category (Paragraph 0127 teaches each event in prior events can include a location (e.g., a street address), a time (event occurrence time), an event category, an event truthfulness, an event severity, and an event description); generating a computed class based on the truth category membership, wherein the computed class associates, for each relevant data-event member of the truth category, the values of the data-event for each category of the computed class (Paragraph 0127 teaches similarly, geo cell entry includes geo cell, lat/lon, streets, businesses, AIs, and prior events; each event in prior events can include a location (e.g., a street address), a time (event occurrence time), an event category, an event truthfulness, an event severity, and an event description)); neutrosophically analyzing the computed class using multi-level regression analysis to identify (a) one or more first level outlier data-event scenarios for each category of the computed class, and (b) one or more next level outlier data-event scenarios for each category of the computed class, so as to determine a plurality of unique multi-level outlier data-event scenarios (Paragraphs 0214, 0263, 0267, and 0270 teach multi-source classifier can implement any of a variety of algorithms including: logistic regression, random forest (RF), support vector machines (SVM), gradient boosting (GBDT), linear, regression, etc; multisource event detection systems can identify (detect) major events and also detect shorter term events within (e.g., a context of) identified (detected) major events; major events can be identified (detected) as anomalies via their characteristics, including Signal Volume, Signal Diversity, Severity, Content, Historical Events, etc.; severity may prove optimally reliable in detecting major events; as depicted, computer architecture includes normalized signal ingestor, signal aggregator, detection classifier, major event handler, major event classifier, notification, signal database, historical major event database, and current major event database; anomaly detection can include a multi-source event detection system “comparing” a current event to past detected events in the same area); and neutrosophically analyzing the plurality of unique multi-level outlier data-event scenarios to identify one or more systemic occurrences of data-event scenarios that do not correspond to the predefined outlier data-event scenarios (Paragraph 0302 teaches some or all of the prior event information can be used to determine whether a current event is an anomaly; an event is classified as “close to” (or potentially) an anomaly; events that are “close to” an anomaly, can be stored and reviewed periodically to determine if classification as an anomaly is appropriate based on the sufficiency of further information).
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Statton and Tibbet to incorporate the further teachings of Tibbet for executing the function in accordance with the redefined domain to comprise: applying one or more rules to the redefined domain so as to determine membership in a truth category; generating a computed class based on the truth category membership, wherein the computed class associates, for each relevant data-event member of the truth category, the values of the data-event for each category of the computed class; neutrosophically analyzing the computed class using multi-level regression analysis to identify (a) one or more first level outlier data-event scenarios for each category of the computed class, and (b) one or more next level outlier data-event scenarios for each category of the computed class, so as to determine a plurality of unique multi-level outlier data-event scenarios; and neutrosophically analyzing the plurality of unique multi-level outlier data-event scenarios to identify one or more systemic occurrences of data-event scenarios that do not correspond to the predefined outlier data-event scenarios.
There is motivation to further combine Tibbet into the combination of Statton and Tibbet because of the same reasons listed above for claim 6.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Stennett et al. (US 20200379974) teaches a method is provided for preventing dark data in a data set. At a time t1, a first version of the data set is received. The first version is analyzed and its parameters are gathered in a first statistical profile. The first statistical profile is stored. At a time t2, a second version of the data set is received. The second version is analyzed and its parameters are gathered in a second statistical profile. The second statistical profile is stored. The first and second statistical profiles are compared and a similarity index is created. If the similarity index exceeds a pre-set threshold, dissimilarity is flagged and a responsive action is taken.
Guan et al. (US 20210042767) teaches a digital content communication system for account management and predictive analytics may be provided. The system may include an analytics system that communicates with one or more servers and one or more data stores to provide digital content management in a network. The analytics system may include a data access interface to receive data associated with a customer, as well as a processor to: standardize the received data using a standardization technique; process the standardized data using a dark data processing technique; generate a customer fit score and a digital density score based on the dark data processing of the standardized data; match received data associated with a customer against at least one variable using at least one matching technique; create a lead analytical record (LAR); prioritize leads in the LAR using a predictive modeling technique; and establish optimized channel assignment based on at least one of the customer fit score, the digital intensity score, the LAR, or the matching and prioritization actions.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to COURTNEY JONES whose telephone number is (469) 295-9137. The examiner can normally be reached on 7:30 am - 4:30 pm CST (M-Th).
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, Neha Patel can be reached at (571) 270-1492. 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 Patent Center. Status information for published applications may be obtained from Patent Center. Status information for unpublished applications is available through Patent Center for authorized users only. Should you have questions about access to Patent Center, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
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) Form at https://www.uspto.gov/patents/uspto-automated- interview-request-air-form.
/COURTNEY P JONES/Primary Examiner, Art Unit 3699
Prosecution Insights