Provided is an information search system by which high-speed search is possible commonly used across a plurality of districts, the system including: a database (12) that stores a plurality of pieces of information that are text-searchable; a query sentence acceptance unit (26) that accepts a query sentence; an inputted search keyword extractor (44) that extracts an inputted search keyword from the query sentence; a retrieval executor (40) that executes retrieval processing from the database using the inputted search keyword; a local management apparatus (100) that stores district material in a local database (104); and an information management apparatus (110) that executes character extraction processing on the material stored in the local database and converts a file format of the material according to a size thereof, stores the material in a temporary memory as stored material, and outputs the stored material to the database.

Techniques for ontology linking of unstructured text as a service are described. A service may receive a request to link unstructured text to a standardized ontology, and the service may segment and tokenize the unstructured text and send the result to multiple services implementing multiple deep machine learning models trained to identify particular entities and one or more relationships between entities. The service may perform a search of the standardized ontology to identify a set of similar candidates from the standardized ontology for the detected entities and the one or more relationships, and then rank the set of similar candidates from the standardized ontology according to their similarity to the detected entities within the unstructured text. The output from the service may include a result identifying a highest ranked candidate of the set of similar candidates from the standardized ontology for the detected entities within the unstructured text.

A candidate document is received, for example, by a document filter. A determination is made based on the content of the candidate document, whether the candidate document is relevant to a document corpus. A determination is made based on the content of the candidate document, whether the candidate document is novel with respect to the document corpus. In response to determining that the candidate document is relevant to the document corpus and novel with respect to the document corpus, the candidate document is added to the document corpus to make at least a portion of the content of the candidate document available for a response to a search query.

Systems and methods are described herein for interpreting natural language search queries that account for contextual relevance of words of the search query that would ordinarily not be processed, including, for example, processing each word of the query. Each term or phrase is associated with a respective part of speech, and a frequency of occurrence of a combination of adjacent terms or phrases public domain is determined. A relevance of each term is then determined based on its respective type of term and frequency of occurrence in the public domain. The natural language search query is then interpreted based on the importance or relevance of each term.

Systems and methods for indexing geological features are disclosed. In one embodiment, a method for indexing geological features includes accessing a database storing a plurality of map objects that originate from documents. Each map object includes a map defined by a geographical boundary and a text caption. The method includes, for each map object, determining a plurality of geohashes within the geographical boundary, and includes, for each map object, comparing terms of the text caption with a list of geological keywords. For each map object, the method includes identifying one or more geological noun phrases within the text caption that match one or more geological noun phrases of the list. The method includes determining, for each geological noun phrase, one or more geohashes associated with the geological noun phrase and, for each geohash, determining a frequency that the geohash is associated with the geological noun phrase.

A training system produces a resource-efficient machine-trained model via a training architecture that employs plural processing paths. Some of the processing paths incorporate the use of auxiliary information that imparts external knowledge about source items being processed. The training architecture also employs contrastive learning that operates at different respective levels within the training architecture. For instance, the training architecture uses encoder-level contrastive learning to compare output information generated by different encoders within the training architecture. The training architecture uses decoder-level contrastive learning to compare output information produced by different decoders within the training architecture. An inference-stage system performs an application task using the model produced by the training system.

Embodiments illustrated herein disclose a method includes receiving a text input, wherein the text input corresponds to a search string. The method further includes converting the text input to a string vector. Additionally, method further includes retrieve, by the processor, one or more phrases in the text input. Further, the method includes predicting one or more technology classifications associated with the text input based on the string vector by utilizing a Machine Learning (ML) model. The method includes generating at least a first structured search string based on the one or more technology classifications and the one or more phrases.

A method includes receiving, in a first query interface, a query composed by the user by typing commands into a query box of the first query interface and based on the receiving of the query, causing events corresponding to query results of the query to be displayed in the first query interface with fields corresponding to the events. Based on the selection by the user of an option, a second query interface is displayed with a table that includes events that correspond to query results of a loaded query. The table includes columns corresponding to event attributes, rows corresponding to events. Cells are populated with the data items of event attributes, where one of the columns corresponds to a field of the fields displayed in the first query interface. The table also includes interactive regions selectable by the user to add one or more commands to the loaded query.

Provided are techniques for data analysis and rule generation for providing a recommendation. Current features are identified from data in a corpus. In response to receiving an indication that the data has changed, a new feature is identified. A feature set is created by identifying one or more related features of the current features. A feature worthiness score for the feature set is generated. In response to the feature worthiness score exceeding a threshold, the feature set is input to a model. One or more rules from the model are received, where each of the one or more rules includes the one or more related features, the new feature, and a recommendation. In response to receiving a set of values for the one or more related features and the new feature, a rule of the one or more rules is applied to provide the recommendation for that set of values.

One example provides, on a computing device, a method comprising iteratively receiving, from one or more collaborative communications applications, user data comprising communications of a user, analyzing the received user data to extract contextual information regarding the user, and sending the contextual information to a contextual information data store for the user. The method further comprises receiving a search query originating from within the collaborative communications application, parsing the search query to identify a referential search term, querying the contextual information data store for the user with the referential search term to identify previously stored contextual information regarding the user associated with the referential search term, and sending the previously stored contextual information identified to the collaborative communications application from which the search query was received.