50 results about "Ranking (information retrieval)" patented technology

System and method for the indexing and retrieval of semantically annotated information units from a collection of semantically annotated indexed information units in response to a query using an ontology-based IR model. The retrieval method comprises: receiving a semantically annotated query with semantic annotations to individuals or classes within a determined populated base ontology; embedding, as a set of weighted-mentions to individuals or classes within the populated base ontology, the semantically annotated query in a semantic representation space of an ontology-based metric space IR model; obtaining the representation in the semantic representation space for every indexed information unit; computing the Hausdorff distance between the space representation of the query and the space representation of all the indexed information units of the collection; retrieving and ranking, the relevant information units based on the computed Hausdorff distance.

The present invention presents an apparatus and system for providing an ability to conduct a secondary search using results provided by a first search capability. This secondary search is integrated with the first search and functions as an added tool or accessory. The present invention allows for user control of search ranking, search viewing and search presentations thus affording more relevant information retrieval. Searchers can dynamically and interactively examine and manipulate the search results to improve relevance and quickly satisfy their search objectives.

A method is provided for retrieving information from massive databases (i.e., databases with millions of documents) in real time, that allows users to control the trade-off between accuracy in retrieved results and response times. The method may be applied to databases with contents, i.e., documents which have been modeled with a clearly defined metric that enables computation of distances between any two documents, so that pairs of documents which are “closer” with respect to the metric are more similar than pairs of documents which are “further apart”. Our method can be applied to similarity ranking and / or can be combined together with other methods to increase the scalability of information retrieval, detection, ranking, and tracking.

The invention discloses a distributed parallel information retrieval system and a distributed parallel information retrieval method. A data acquisition unit acquires data from the internet; an index unit processes and organizes the acquired data to establish a data structure and a storage mode which are suitable for retrieval; and a retrieval unit analyzes and processes query results which are submitted by a user, carries out similarity calculation and ranking, collects the query results and carries out mass data retrieval by using parallel processing and distributed processing technologies.

An information retrieval system for scientific literature provides a summary document for reviewing the results of a query that provides both a text-based summary and an image-based summary, the latter represented by a figure from the document that has a highest ranking indicating its centrality to the concept of the document. A ranking of figures may be done by an analysis of figure-associated text either by supervised machine learning or comparison of text similarities.

Disclosed herein is a system and method of using the same for a relevancy ranking information retrieval system. In an embodiment, the system is configured for ranking hits in text string searching. A search query for one or more hits relevant to an investigation is received from, as an example and not a limitation, a user. A set of attributes and features of each attribute are extracted related to metadata for each of the one or more hits. A score of each attribute is calculated based on the metadata features, although not limited to ‘metadata’ information as typically defined in digital forensics. Further, weights are assigned to each of the one or more attribute features and a relevancy rank is generated for each of the one or more hits based on assigned weights and the attribute score by using a predefined relevancy-ranking algorithm that may be adjusted by user input.

The invention provides a multi-modal heterogeneous model retrieval enhancement method and system, and the method comprises the steps: building a knowledge and application example double-corpus based on user multi-modal query, and designing a joint retrieval mechanism to obtain a result set; mapping and scheduling to obtain feature representation through special processing channels for texts, images and audios and a Spiking neural network with a segmented trapezoidal topological structure; constructing a three-stage cascade architecture of a basic model, an advanced model and human experts, and obtaining a decision path and answer candidate set in combination with a recursive and discarding decision mechanism; a Hamiltonian graph network is used for representing a multi-modal relation, and a gradient-free descent method is used for rapidly training and optimizing model parameters; an enhanced retrieval result is obtained through cross-modal semantic alignment and dynamic retrieval window adjustment; and high-quality response is obtained through context-aware sorting and retrieval enhanced reasoning. According to the method, the multi-modal information retrieval processing efficiency and the heterogeneous model reasoning response quality are improved.

Provided are systems, methods, and apparatuses for systems and methods of memory efficient multi-vector information retrieval based on embeddings from a storage pipelined network. In one or more examples, the systems, devices, and methods include performing a first portion of a nearest neighbor search on a first subset of a set of media sources; performing a fetch process on the first subset that include identifying a first set of highest matching media sources from the first subset, transferring multi-vector representations of the first set of highest matching media sources from a storage drive to a memory, and performing a second ranking of the first set highest matching media sources. The systems, devices, and methods include performing, in parallel with the fetch process, a second portion of the nearest neighbor search on a second subset of the set of media sources.

An information retrieval system for scientific literature provides a summary document for reviewing the results of a query that provides both a text-based summary and an image-based summary, the latter represented by a figure from the document that has a highest ranking indicating its centrality to the concept of the document. A ranking of figures may be done by an analysis of figure-associated text either by supervised machine learning or comparison of text similarities.

The embodiment of the invention relates to the technical field of data processing, and particularly provides a smart library information retrieval method and system based on AI session interaction. According to the embodiment of the invention, an information retrieval session request initiated by a user through a smart library interaction interface is received; collecting session interaction data containing natural language query content and context association information; performing semantic structuring processing on the session interaction data to generate a session semantic feature set containing an entity relationship network and an intention classification vector; carrying out deep mining on the session semantic feature set by utilizing a debugged retrieval intention understanding model, identifying potential retrieval demands and demand priority ranking, and generating retrieval demand feature vectors; and on the basis of library collection information in the vector matching intelligent library resource database, generating an information retrieval result set containing resource association degree sorting, and feeding back the information retrieval result set to the interactive interface for visual display. According to the embodiment of the invention, the accuracy and comprehensiveness of information retrieval are improved, and more intelligent and efficient retrieval experience is provided for users.

Systems and method for generating a response to a query. The method includes using a first large language model (LLM) to generate synthetic information related to a query; generating an amended query based on the synthetic information related to the query; using an information retrieval system to retrieve, from a plurality of chunks, a set of chunks that are relevant to the amended query, wherein each chunk of the plurality of chunks is all or a portion of a document in a corpus of documents; using a second LLM to rank the set of chunks based on a relevance to the query; selecting a subset of chunks from the set of chunks based on the ranking; and using a third LLM to generate a response to the query based on the subset of chunks.

The invention discloses a self-adaptive multi-feature fusion hybrid retrieval sorting method and system, and belongs to the technical field of information retrieval. The method comprises the steps that after user query is received, a mixed retrieval process and an intention recognition process are executed in parallel; performing multi-dimensional feature extraction on the candidate documents obtained by the mixed retrieval, wherein the multi-dimensional feature extraction comprises semantic correlation features, keyword matching features, document authority features and timeliness features; and according to the identified query type, adaptively selecting a fusion weight, and carrying out weighted fusion on the multi-dimensional feature vector to calculate a final score and sort the final score. According to the method, the problem that weight distribution is rigid in traditional mixed retrieval is solved through an intention self-adaptive dynamic weight mechanism, meanwhile, by introducing multi-dimensional service features, the ranking result not only ensures the correlation, but also meets the quality requirement under a service scene, and the accuracy and practicability of a retrieval system are remarkably improved.

The invention discloses a method and a system for performing multi-dimensional combined retrieval on enterprise information. The method specifically comprises the following steps: 1, constructing a foundation; 2, carrying out learning perfection; 3, carrying out searching and displaying; 4, carrying out searching and optimizing. The invention relates to the technical field of information retrieval. According to the method and the system for performing multi-dimensional combination retrieval on the enterprise information, enterprise related information data is collected through the Internet, a dictionary is constructed by an NLP related technology, a bert + bolstm + crf model architecture is adopted to learn a perfect recognition method, a technical framework based on a fasti python micro-service framework + docker + NLP and deep learning is formed, enterprise information is identified by utilizing an intelligent identification algorithm, and retrieval service is provided in a multi-dimensional combined retrieval mode, so a large number of retrieval skills are not needed, the requirements of a large number of users can be met, the experience feeling of the users is improved, the website jumping rate is reduced, and a positive help is provided for website ranking.

An example system includes a processor to receive production log data and pre-process the production log data to generate an exploration data set. The production log data includes continuous updates of a ranking function associated with different actions for a same query issued by a same user in different impressions. The processor can perform offline estimation of online metrics for ranking functions using the exploration data set, evaluate a proposed ranking function by comparing the estimated online metrics to a set of baseline metrics of a baseline ranking function, detect that the proposed ranking function is a preferred ranking function having a higher detected metric score than the baseline ranking function based on a comparison of the estimated online metrics with the baseline metrics of the baseline ranking function, and execute an action on an information retrieval system based on the preferred ranking function.

Disclosed implementations for detecting exploitation of ranking signals used to provide search results. An expected value for a ranking signal is determined based on a plurality of resources responsive to a query. A residual value is determined by aggregating a difference between the expected value and an information retrieval score for the ranking signal across a domain, wherein the domain includes at least one of the plurality of resources. Responsive to determining the residual value is indicative of an exploit, adjust a ranking of a resource associated with the domain in a search result page, the resource responsive to a second query based on the ranking signal.

The invention provides a retrieval enhancement generation system document reordering method based on big language model confidence, electronic equipment and a storage medium, and belongs to the field of information retrieval. According to the method, it is assumed that the confidence coefficient of LLM generation response is positively correlated with the correlation of input documents, on this basis, for a given query and document set, document ranking is optimized through an LLM confidence coefficient signal, and the ranking performance is improved in combination with a threshold mechanism and a multi-level ranking strategy. According to the method, an additional training model is not needed, calculation is efficient, an efficient and generalizable solution can be provided by utilizing the lightweight pre-training LLM of the 7-9B parameter, the accuracy and efficiency of document reordering are remarkably improved, the illusion phenomenon is reduced, the overall reliability of knowledge-intensive tasks (such as questions and answers and fact verification) is improved, and the method is suitable for popularization and application. The performance of the retrieval enhancement generation system is remarkably improved, and the method is suitable for various retrievers and resequencers.

The invention discloses an AI-based geological GIS model rapid retrieval method and system, and relates to the technical field of geological information retrieval, and the method comprises the following steps: obtaining a geological GIS model library, extracting geological attribute characteristic parameters and unique identifiers in the geological GIS model library, forming a mapping relation, forming a relation mapping table, and obtaining geological data to be recognized. Geological attribute analysis is carried out on the to-be-recognized geological data to obtain to-be-recognized geological attribute parameters, first retrieval is carried out on the to-be-recognized geological attribute parameters based on the relation mapping table to obtain a candidate set, second retrieval is carried out on the candidate set to obtain a comprehensive matching degree, and the geological GIS models in the candidate set are subjected to priority ranking according to the comprehensive matching degree to obtain the geological GIS models. And outputting the retrieval results sorted according to the priorities. The problems that a traditional method is low in efficiency and insufficient in precision are effectively solved, and the efficiency and accuracy of geological GIS model retrieval are remarkably improved.

An apparatus allows the semi-automated creation and curation of a knowledge graph based on a query-atom ER ranking engine. This apparatus cooperates with a domain expert to smoothly and semi-automatically incorporate or restructure textual data in the knowledge graph when a suitable high-confidence response to a query cannot be found. The apparatus extends the conventional information retrieval approach to consumer interaction by building a structured knowledge graph. Using graph exploration, the apparatus augments the ranking made by the underlying neural model in order to stay in sync with the constantly-changing domain of application. The apparatus allows for the consistent, cognitive, consumer-driven incorporation and restructuring of relevant unstructured data in a knowledge graph which is generated by mimicking users' behaviors.

The invention provides an object sorting method and device, a model training method and device, a medium and equipment, and relates to the technical field of artificial intelligence, in particular to the technical field of machine learning and information retrieval. According to the implementation scheme, the method comprises the steps of obtaining a candidate list, a current search word corresponding to the candidate list and a historical behavior sequence; based on the historical behavior sequence and the current search word, utilizing a first interaction network in a sorting model to obtain a first fusion feature; for each candidate object in the candidate list, based on the object information of the candidate object and the first fusion feature, predicting to obtain a prediction score of the candidate object; and sorting the candidate objects in the candidate list based on the predicted score of each candidate object in the candidate list.

The invention discloses a lengthy query thumbnail method fusing key concepts and potential concepts, and relates to the field of natural language processing. Before information retrieval is carried out, lengthy query input is abbreviated. Key concepts representing query core content are combined with potential concepts that are important to understand the query but not explicitly expressed to generate a more complete and effective query. Specifically, firstly, a short and effective query is generated through fine tuning of a T5 model, and a short query serving as a key concept is generated according to semantic content of an input text; secondly, mining potential concepts of the original query by using a pseudo-correlation feedback method, and selecting n terms with the highest inverse document frequency in k documents ranked in the top in the initial retrieval result as the potential concepts of the original query; and finally, aggregating the two as a final query thumbnail result to realize lengthy query retrieval. According to the method, the recall rate and the accuracy of the retrieval result are improved.

The invention discloses a product information retrieval method for automobile vertical websites. The method includes the steps that multiple pieces of automobile vertical website data are captured with the crawler technology and are stored in database according to a website-automobile type-public praise tree structure; data formats in the database are normalized and preprocessed to make data formats and naming ways from different websites unified; by considering the websites from which the data comes, public praise grades, public praise content and user grades, a retrieval result evaluation system is constructed, and an index weight is allocated to the retrieval result evaluation system; a public praise feature vector is constructed according to the retrieval result evaluation system, andin combination with a sentiment analysis score, an automobile product information retrieval result is obtained by integrating the websites to which public praise belongs, the public praise grades andthe user grades of issued information. The method solves the problems that in a traditional retrieval method, the returned result is monotonous, the method adapts to a limited number of new words, andpersonalized ranking of specific users or specific needs can not be achieved.

The invention relates to the technical field of medical information retrieval, and discloses a medical information retrieval method based on an intelligent medical science popularization platform. The method comprises the following steps: acquiring a user retrieval request, and dividing the user retrieval request into a medical concept entity and an association description entity; based on the medical concept entities, extracting core concept nodes and adjacent extension nodes in a preset medical knowledge graph, and generating an initial retrieval range; based on the association description entity, performing semantic filtering on the initial retrieval range, screening out semantic matching nodes, and constructing a target retrieval set; analyzing the medical information in the target retrieval set to generate a medical information optimization evaluation value; and performing priority ranking on the target retrieval set according to the evaluation value, and outputting a ranked medical information retrieval result. According to the method, the matching degree of the retrieval result and the user demand is improved, the user can conveniently and quickly obtain high-quality medical information, the retrieval experience is improved, and the method is suitable for an information retrieval scene of an intelligent medical science popularization platform.

An example system includes a processor to receive production log data and pre-process the production log data to generate an exploration data set. The production log data includes continuous updates of a ranking function associated with different actions for a same query issued by a same user in different impressions. The processor can perform offline estimation of online metrics for ranking functions using the exploration data set, evaluate a proposed ranking function by comparing the estimated online metrics to a set of baseline metrics of a baseline ranking function, detect that the proposed ranking function is a preferred ranking function having a higher detected metric score than the baseline ranking function based on a comparison of the estimated online metrics with the baseline metrics of the baseline ranking function, and execute an action on an information retrieval system based on the preferred ranking function.

The invention provides an enhanced spurious correlation feedback model information retrieval method based on a hyperspace simulation language. The method includes the steps of retrieving and feeding back first N documents according to query words Q of an initial user, using all words in the first N documents of the initial feedback as candidate extended words, and adopting a hyperspace simulationlanguage model for calculating adjacent word frequency weights according to the position distance between the candidate extended words and the initial query words; calculating adjacent word weights ofthe final candidate extended words and all the query words; conducting fusion on the adjacent word weights of the candidate extended words and importance weights of the candidate extended words in spurious correlation document sets to obtain final query word items, and obtaining a final feedback document ranking. The accuracy of information retrieval results is further improved effectively, required information are provided move quickly and better for users, searching time and resource costs are saved, and generation of the method has a significant meaning.

The invention provides a database scene-oriented information retrieval method and device and electronic equipment, and relates to the technical field of data processing, in particular to the technical field of information recommendation. According to the specific implementation scheme, the method comprises the steps of receiving a recall request, and performing information recall at different data nodes according to the recall request to obtain at least one piece of recalled candidate information; obtaining node rankings of different data nodes; and performing ranking optimization on the candidate information based on the node ranking, and feeding back target information to the client according to the optimized ranking of the candidate information.

A method is presented for enhancing search results by segmenting documents into smaller passages and utilizing those passages as the search unit. This method integrates the passage rankings from two search models to produce a new document ranking and arranges the documents accordingly. The ranking of document is also rearranged based on the proportion of passages taken from the same document versus the total number of passages in that document. The final search ranking system combines document-level and passage-level search rankings to rank documents. This method ensures that when conducting a passage search, the search results consider the general relevance of the entire document, which leads to better performance compared to searching only by document-level or passage-level searches.

Provided are systems, methods, and apparatuses for systems and methods of memory efficient multi-vector information retrieval based on embeddings from a storage pipelined network. In one or more examples, the systems, devices, and methods include performing a first portion of a nearest neighbor search on a first subset of a set of media sources; performing a fetch process on the first subset that include identifying a first set of highest matching media sources from the first subset, transferring multi-vector representations of the first set of highest matching media sources from a storage drive to a memory, and performing a second ranking of the first set highest matching media sources. The systems, devices, and methods include performing, in parallel with the fetch process, a second portion of the nearest neighbor search on a second subset of the set of media sources.

The invention provides a public opinion information retrieval method based on bidirectional enhancement of a big language model and a knowledge graph. The method comprises the following steps: constructing a bidirectional enhancement mechanism, enhancing the big language model through the knowledge graph, and enhancing the knowledge graph through the big language model to obtain an enhanced big language model and the enhanced knowledge graph; acquiring real-time public opinion data in real time, and dynamically constructing a knowledge graph by adopting the bidirectional enhancement mechanism according to the real-time public opinion data; personalized retrieval and sorting are carried out, retrieval calculation is carried out on the knowledge graph and the content in the public opinion data retrieval pool according to user retrieval content, a retrieval result is obtained through sorting, and public opinion information retrieval based on bidirectional enhancement of the large language model and the knowledge graph is completed. According to the method, the crossing from a static knowledge base to a dynamic cognitive system can be realized, and second-level response knowledge reserve and deduction capability are provided for decision making; the qualitative change from a retrieval result to decision insight is realized, the user is effectively enabled, and the user stickiness is improved.

The invention discloses a keyword similarity ranking method and system, and belongs to the field of information retrieval, and the method comprises the following steps: 1, synchronizing heterogeneous data sources; 2, data preprocessing and conversion, wherein a deep learning model is adopted to conduct automatic preprocessing and conversion on heterogeneous data of different formats; step 3, real-time data synchronization: performing synchronization processing on the real-time data stream by using stream-oriented computation and a real-time data processing framework; step 4, dynamic keyword weight calculation; 5, keyword similarity ranking is carried out, semantic understanding and similarity calculation are carried out on the keywords through a large model, and optimized keyword ranking is generated in combination with dynamically calculated keyword weights; and 6, performing personalized ranking and recommendation. According to the method, the performance is excellent, the expansibility is high, flexible weight adjustment can be carried out when dynamic data are processed, and then the accuracy of a retrieval result is effectively improved.