771 results about "Search engine" patented technology

The invention discloses a multi-modal bill processing method based on dynamic knowledge enhancement. The multi-modal bill processing method comprises the following steps: S1, constructing a dynamic knowledge base containing an aging weight; s2, synchronously processing text, image and format features of the bill by adopting a multi-modal feature fusion network to generate a composite feature vector; s3, semantic-level, format-level and timeliness three-level fusion retrieval is carried out based on the composite feature vector, and a three-level fusion retrieval engine comprises dynamic weighted sorting with timeliness attenuation, a difference degree triggered artificial review mechanism and a policy sensitive slope adjustment algorithm; s4, setting a multi-expert cooperative verification system, wherein the multi-expert cooperative verification system comprises cooperative work of a rule engine, a large language model and a logical reasoning module; s5, implementing a dynamic knowledge updating mechanism, and automatically triggering incremental learning of the knowledge base when policy change or format update is detected; and S6, outputting structured data, and synchronously generating an auditing traceability chain containing a decision path. According to the method, the key field identification accuracy can be improved, and auditing traceability and non-perceptual increment updating in the whole process are realized.

The invention discloses an intelligent recommendation method for optimizing advertisement keyword combination through cross validation, and relates to the technical field of advertisement technology and search engine marketing, which comprises the following steps: constructing a heterogeneous data set through multi-modal data fusion, and layering according to data sparseness: training a Transform-XL time sequence model by adopting time cross validation of a dynamic K value in a high resource layer; a graph neural network association graph is introduced into a low resource layer, semantic expression of a long tail word is enhanced, a stratified sampling-transfer learning two-channel mechanism is designed, and the generalization ability is improved in combination with exposure frequency weighting and a parameter freezing strategy; developing a Bayesian fusion engine, and dynamically weighting a high / low resource layer prediction result by using an improved Materon kernel function Gaussian process; and generating a confidence interval based on neural quantile regression, and outputting an optimal keyword combination sequence under ROI-risk-diversity constraint in combination with multi-target Pareto optimization. According to the method, the cold start efficiency and the long-tail resource utilization rate are improved, and high-robustness decision support is provided for advertisement putting.

The invention relates to the technical field of natural language processing, in particular to an intelligent search engine system and method based on NLP and vector hybrid retrieval, and the system comprises a query analysis module which is used for receiving a query statement input by a user and obtaining structured query information based on the query statement; the symbiotic index module comprises a sparse extension index unit which is used for constructing a sparse inverted index table based on keywords in the structured query information; the vectorization hypergraph index unit is used for forming a hypergraph index based on the business data; the recall arrangement module is used for acquiring a candidate object set according to the structured query information; the constraint rearrangement module is used for calculating a joint priority score between query and candidate objects based on the candidate object set, and obtaining a sorting result under constraint conditions of meeting a supplier proportion, a category proportion and a price interval; and the evidence generation module is used for generating evidence information based on the sorting result and outputting the evidence information to the user interface.

An augmented search engine is provided. The augmented search engine receives an initial query for a search from a user. The augmented search engine determines a next search phase using a search state database. When the next search phase includes requesting additional user input, the augmented search engine generates a user prompt using the initial query, displays this prompt to the user, and stores the prompt in the search state database. Upon receiving user input in response to the prompt, the augmented search engine generates a search query using the initial query, the user prompt, and the user input and determines search results by querying search indexes with the search query and stores these results in the search state database. The augmented search engine generates a search summary using the initial query, the user prompt, the user input, and the search results, and displays this summary to the user.

In various examples, systems and techniques are provided that encapsulate indexing and query operations into an application programming interface (API) that automates and coordinates calls to various local and cloud-based services. When a user has a document(s) to add to a retrieval augmented generation (RAG) database, the API may offer to the user multiple document processing pipelines (DPPs) having pre-set indexing configurations. Similarly, when a user query is received, the API may generate calls to implement query processing that does not require the user to manually configure retrieval and processing of the embeddings. The API may further implement calls that locate a relevant embedding store and provide the stored embeddings, together with the query embeddings, to a search engine that identifies the most relevant matches. The API may then access the embedding-to-text indexing and identify relevant text segments and documents to a prompt generator.

The invention discloses an intelligent marketing system based on user portraits and behavior prediction, which belongs to the field of artificial intelligence, and comprises multi-source data acquisition and integration: a diversified data acquisition interface is configured, and user multi-type data can be acquired from multiple channels such as an e-commerce platform, social media, a search engine and an enterprise CRM system; comprising basic information, browsing and purchasing behavior records, social interaction information, search preferences and emotion-related text data. Data of different formats and sources are integrated into a unified data set by using preprocessing means such as data cleaning, duplicate removal and standardization; according to the invention, through accurate user portrait and behavior prediction, an enterprise can accurately grasp the demand preference of a target customer, accurate pushing of marketing information is realized, the hit rate and the conversion rate of marketing activities are greatly improved, and the marketing effect is effectively improved; according to user value and behavior prediction, marketing resources are reasonably allocated, resource waste is avoided, and enterprise marketing efficiency is improved.

The invention provides an intelligent environment damage questionnaire generation and semantic analysis method and system, and the method comprises the steps: obtaining structured environment data through a government department interface, and capturing network public data in combination with an AI search engine; performing cleaning and normalization processing on the data, and constructing a standardized environment database; constructing a semantic model based on an NLP technology, dynamically generating a customized investigation problem according to an interview object type and an event scene classifier, and optimizing the priority of the problem; extracting a key entity from the interview text by adopting a pre-training model; mapping the entity to a preset pollution category through the environmental damage ontology library, and generating a dynamic causal association map; and verifying the integrity and consistency of the atlas by using a logic verification rule, and automatically outputting a structured survey report. According to the method, the defects that a traditional questionnaire is rigid in design, fragmented in information and insufficient in implicit association mining are overcome, the accuracy and efficiency of environmental damage investigation are remarkably improved, and intelligent support is provided for pollution traceability and responsibility definition.

The invention relates to an information retrieval method and device for hierarchical planning reinforcement learning based on retrieval enhancement. The method comprises the following steps: acquiring a natural language query request of a user; calling a large language model of hierarchical planning reinforcement learning training based on retrieval enhancement to generate semantic keywords; the high-level strategy splits the natural language query request into a series of sub-query requests; the low-level strategy generates a retrieval query according to the context of the current sub-query request, and obtains a result verification condition from an external knowledge base through an RAG model; calling a search engine to execute multiple rounds of fine-grained information retrieval to obtain a plurality of candidate retrieval results; and performing deep semantic matching analysis on the candidate retrieval results and the result verification conditions by utilizing the post-trained large language model to calculate a matching degree score, and screening the candidate results according to the matching degree to obtain a final information retrieval result. Compared with the prior art, the method has the advantages that high-quality content retrieval and result display of complex query problems of users in various scenes can be realized.

Methods and systems for interacting with users via a chatbot. A natural language query is received and processed by submitting a search query to a search engine. The search engine identifies relevant information including textual information and images for formulating a response. The identified information and query are submitted to a Large Language Model which generates a response displayed via the chatbot. The response may include textual information and relevant images. The system can extract text from images of documents and convert textual information into numerical vector representations for processing. Selectable options based on clustered relevant information can be provided to users for query refinement when appropriate. The chatbot interface enables natural language interactions while leveraging search capabilities and Artificial Intelligence to provide informative and helpful responses with both text and visual elements.

This disclosure describes utilizing a generative document system to dynamically build and provide generative text documents using one or more generative artificial intelligence (AI) models. For example, the generative document system efficiently utilizes various systems and one or more generative AI models to determine intents and topics, curate topic sections, and generate a generative text document that includes a directed answer along with select curated topic sections for search queries. In various implementations, the generative document system performs additional actions that enhance the efficiency and accuracy of operations used to produce generative text documents. Additionally, in many cases, these generative text documents provide a foundation for providing an interactive, intuitive, wide-ranging, and flexible curation of answers to users that address the corresponding search queries.

The invention relates to the technical field of text generation, provides an NL2SQL method and system based on thinking reasoning, and aims to enable a model to interact with a database environment by introducing a reinforcement learning mechanism to obtain reward signal feedback. The system comprises a table field recall component and a structured query language generation component. The table field recall component fills table field information into a preset template to construct a structured text, then vectorization processing is carried out, the structured text is stored into an elastic search engine, a relevant table is searched according to user query, and finally specific fields are screened through a field selector model. And the structured query language generation component receives the screening result, generates a structured query language in combination with user query, and performs retrieval replacement on a filtering value. In the aspect of training, the method combines supervised fine-tuning cold start and reinforcement learning reasoning training, and the model performs reasoning and thinking according to an input context through trial and error learning. In the reinforcement learning training process, a reward function iteration optimization model based on task execution result correctness is adopted.

Systems, methods, devices, and computer readable storage media described herein provide techniques for intelligently interpreting and / or responding to search queries. In an aspect, a search query comprising a search term is received. A generative artificial intelligence (AI) model is utilized to generate a question based on the search term. Data semantically similar to the question is identified. The generative AI model is used to determine an answer to the question based on the question and the identified data. In a further aspect, a question-answer pair comprising the question and answer is generated. In an alternative further aspect, the answer is caused to be presented in a graphic user interface corresponding to a search engine. In a further aspect, the answer is determined during an active session with the search engine. In a further aspect, the search term and the question-answer pair are stored in a key-value store.

Disclosed is a computer-implemented method for constructing a domain-specific language learning model (LLM). The computer-implemented method includes a step of ingesting, from a server, a domain-focused dataset. The computer-implemented method includes a step of assimilating, from a search engine database, a real-time digital data stream. The computer-implemented method includes a step of integrating the domain-focused dataset, and the real-time digital data stream within an application logic layer to obtain an integrated dataset. The computer-implemented method includes a step of employing, by a processor, a transformer algorithm on the integrated dataset to extract a plurality of domain-specific textual insights. The computer-implemented method includes a step of utilizing, by the processor, said domain-specific textual insights to execute one or more business tasks. The computer-implemented method includes a step of presenting both the domain-specific textual insights and the executed business tasks on a user interface.

The disclosure is directed to various ways of improving the functioning of computer systems, information networks, data stores, search engine systems and methods, and other advantages. Among other things, provided herein are methods, systems, components, processes, modules, blocks, circuits, sub-systems, articles, and other elements (collectively referred to in some cases as the “platform” or the “system”) that collectively enable, in one or more datastores (e.g., where each datastore may include one or more databases) and systems, the creation, development, maintenance, and use of a set of custom objects (and core objects) for use in a wide range of activities, including sales activities, marketing activities, service activities, content development activities, and others, as well as improved methods and systems for sales, marketing and services that make use of such entity resolution systems and methods as well as custom objects.

Techniques for determining whether retrieved content is sufficient for a language model to generate a response to an input are described. In some embodiments, a system may determine a retrieval complexity (RC) metric based on a question, a reference answer and document results retrieved from a retrieval component (e.g., a search engine, a knowledge search, etc.). The RC metric may be based on whether one of the document results corresponds to the reference answer and whether the reference answer can be determined from the set of the document results (e.g., using two or more document results). The RC metric can be used to determine when retrieved content is to be used by a language model for responding to an input corresponding to the question.

An adaptive random access system and method with learned query optimization for compacted data files that enhances random access performance through machine learning and pattern recognition. The system incorporates a query pattern learning module that analyzes historical access patterns and user behavior to build statistical models of data usage. An adaptive estimator module improves location estimation accuracy by incorporating learned patterns rather than relying solely on mathematical calculations. A predictive boundary detector uses learned codeword patterns to more accurately identify boundaries in compacted data, reducing misalignment errors. An intelligent search engine coordinates optimization strategies including context-aware search string parsing and encoding strategy selection based on learned performance data. A dynamic codebook optimizer reorganizes sourceblock layout based on access frequencies and co-occurrence patterns to improve retrieval speed. An enhanced search cache implements predictive caching algorithms that anticipate user queries and proactively load relevant data.

The invention provides an automatic SEO setting system and method in a station building CMS, and belongs to the technical field of station building CMS. According to the method, the AI technology is utilized, the keywords are deeply mined and screened, the keywords with high conversion potential and the long tail keywords are accurately recognized, the real demands of users are insighted, the competition situation of the keywords is intelligently evaluated, and the popularity change is predicted. In the aspect of content creation, a content outline, a first draft and a specific paragraph can be generated according to a target keyword and a user portrait, optimization suggestions can be provided in the aspects of readability, keyword layout and the like, and content personalized recommendation is achieved. Compared with a traditional SEO technology, the method has the advantages that the keyword screening efficiency and the long tail keyword coverage rate are improved, the matching degree of content and user requirements is optimized, ranking and flow of websites in a search engine are improved, and the method is efficient and accurate.

A text processing method includes obtaining a query text, invoking a search engine interface based on the query text to obtain a plurality of text search results corresponding to the query text, obtaining, from the plurality of text search results, a plurality of answer text segments matching the query text, determining a relevance between the query text and each of the plurality of answer text segments, determining one of the plurality of answer text segments that corresponds to a maximum relevance as a reference text of the query text, and invoking a language model based on the query text and the reference text to obtain a reply text of the query text.

A computer-implemented method estimates body dimensions and provides clothing recommendations using frontal and side images from a user device. The images undergo validation via a deep neural network, ensuring correct posture, visibility, and lighting. A convolutional neural network (CNN) extracts geometric features, while segmentation techniques remove background noise. A simplified 3D body model is generated, and key body dimensions—such as shoulders, chest, waist, and legs— are calculated using object detection models and a standardized height reference. The method then suggests clothing sizes by matching the user's measurements with predefined size charts and enables a virtual fitting room for trying on clothes. An auto-tagging system identifies clothing attributes and links them to an image search engine for recommendations. All images are encrypted and securely stored, with an option for permanent deletion.The method can be integrated into third-party applications via an SDK or API, allowing real-time body measurement and virtual fitting functionality.

The invention discloses a knowledge graph-based search engine relation chain recommendation method, which comprises the following steps of: receiving a relation chain recommendation request submitted by a user through a search engine to obtain initial query information; performing semantic analysis and intention recognition on the initial query information, and determining a corresponding entity type and a relationship type; calling a dynamic knowledge selection and evaluation mechanism according to the entity type and the relationship type, and dynamically screening the original knowledge graph to obtain effective knowledge sub-graphs; performing diffusion type reasoning based on a knowledge graph diffusion model to generate a candidate relation chain path set; evaluating the quality score of each candidate path in real time; and performing dynamic screening according to the path quality score, and outputting a relation chain path recommendation result. According to the method, accurate mining and dynamic optimization of the knowledge graph relation chain path are realized, and the accuracy of a recommendation result and the flexibility of user interaction are improved.

An RAG auxiliary generation type search engine method based on multiple rounds of expert and self-adaptive mixed retrieval comprises the steps that a crawler is used for crawling data, the crawled data are processed, a graph knowledge base is constructed through text partitioning and entity relation extraction, the graph knowledge base is coded into semantic vectors, and a vector database is established; by calculating similarity between subject vectors and literature vectors in a vector database, retrieving related literatures, utilizing a large language model to simulate a multi-expert role to generate differentiated viewpoints, and combining multiple rounds of dialogues and user feedback, generating a multi-view fused article to complete multiple rounds of expert answers; designing a double-layer retrieval mechanism according to different nodes and relationships in the retrieval atlas based on the atlas knowledge base; meanwhile, dense retrieval of a vector knowledge base of text coding is combined and utilized, a map-vector collaborative retrieval mechanism is designed, a self-adaptive mechanism and a session memory function are introduced, and multi-round question and answer response is carried out; according to the method and the system, higher logical contents can be generated.

The invention discloses a biochemical experiment automation script training generation method based on RAG and DPO. The method comprises the steps that biochemical experiment technical documents are collected to serve as an external knowledge base; the BM25 and the Fiss are fused to construct a hybrid retriever; generating experimental process description by using a large language model, determining equipment, protocols and materials, and retrieving related documents from a knowledge base on the basis of the experimental process description, the protocols and the materials; generating an experiment script based on a retrieval result, and performing platform simulation verification, and performing iterative optimization on the script which fails in verification; marking successful and failed scripts as preference data pairs, and constructing a training set; carrying out LoRA fine tuning on a local large language model by adopting direct preference optimization; and according to a target and appliance prompt input by a user, generating a verified experiment script by using the large language model after direct preference optimization training. According to the method, the script generation automation level and the passing rate are remarkably improved, the manual intervention cost is reduced, and an efficient solution is provided for automation of biochemical experiments.

The invention relates to the technical field of document processing, and discloses a document knowledge retrieval method and system, electronic equipment and a storage medium. Based on a language model, synchronously generating abstract contents and a semantic association query set for each text fragment, and aggregating the abstract contents of all text fragments corresponding to the same document to form a fragment abstract set; constructing a multi-path retrieval pool based on the text fragment, the abstract content and the semantic association query set, and respectively constructing a plurality of independent retrieval data sources; the method comprises the following steps: constructing a dual-granularity abstract index system, receiving a query request of a user, executing query routing processing based on dual-granularity abstract index and a hierarchical retrieval process based on a metadata structure, returning retrieval result data, performing fusion calculation and duplicate removal processing on the retrieval result data recalled by multiple paths, and obtaining a retrieval result. And meanwhile, the document abstract short sentence group is injected into a search suggestion function of a search engine. According to the method, the retrieval accuracy can be improved, and the user experience is improved.

The invention discloses a unified search and page configuration rendering method and system based on a data model. The method comprises the following steps: S1, data acquisition and model mapping; s2, indexing and synchronizing data; s3, unified search request processing; s4, processing and returning a result set; and S5, dynamically rendering the page. The system comprises a data source layer, a data unification layer, an index layer, an application service layer and a display layer. The invention provides a set of end-to-end integrated solution for solving the problems that multi-source heterogeneous data is difficult to integrate, the cross-system search efficiency is low, page rendering lacks flexibility and the like. According to the scheme, by constructing a standardized data model, a unified search engine index system and a dynamic page rendering mechanism driven by configuration, a complete technical link from data source access to front-end display is achieved, and the efficiency and expansibility of the system in the aspects of data management, information retrieval and page display are remarkably improved.

The invention relates to a dynamic advertisement putting method and system oriented to a large language model, and belongs to the technical field of generative artificial intelligence and calculation advertisement crossing. According to the method, the semantic splitting problem of advertisements and natural language streams is solved through a three-level collaborative architecture, firstly, a semantic depth alignment module is constructed, and advertisement and natural language vectors are projected to a unified semantic space by using dual-channel adversarial training and orthogonal mapping constraints; a dynamic context adaptation module is constructed, dominant intentions are analyzed based on a self-attention mechanism, and scene advertisements are generated by fusing space-time-equipment signals; and finally, constructing an anti-fragility regulation and control module, and dynamically balancing the advertisement density and the user experience through an exponential attenuation model and a reinforcement learning fusing mechanism. The method breaks through the limitation of a static advertisement template, realizes the adaptive fusion of the advertisement content and the generative interaction scene, and is suitable for scenes such as a multi-language dialogue system and an intelligent search engine.

Disclosed are systems and methods that provide a decision-intelligence (DI)-based, computerized framework for compiling and leveraging reliable sequence taggings for input queries related to executed searches. The disclosed framework can compile a trained computer model to fulfill partially labeled queries tagged by AI models as fully labeled queries. The disclosed framework can further leverage other AI models (e.g., deep neural networks, knowledge graphs, and the like), so that cross-checks can be performed between different models to guarantee high quality of labeled tokens. Thus, the framework can automatically generate and implement reliable training data to train a sequence tagging model for search query understanding. Thus, the search engine operating on such tagging model can provide improved results.

The invention discloses a business log analysis method and device, electronic equipment and a storage medium, and relates to the technical field of log analysis. Generating target log data containing a target field according to related log data of each transfer node in the business creation transfer process, the target field comprises an identification field of the current transfer node, an identification field representing an association relationship between the current transfer node and other transfer nodes in a business creation transfer process, and a target log content field corresponding to the current transfer node; creating an index for the target field in a search engine; and when the query service for the target field content is received, the queried target log content associated with the target field content is aggregated and fed back to the query end, so that the problems that the log data of each node is dispersed, and the related log data cannot be quickly analyzed and then quickly positioned are solved.

The invention discloses a data model establishment method based on a large language model and a local knowledge base. The method comprises the following steps: S1, collecting parallel corpora through a distributed crawler technology; s2, converting data in a non-text format into data in a text format; s3, cleaning and preprocessing all the collected data; s4, denoising, segmenting and aligning the content to construct a parallel corpus; s5, storing the processed parallel corpora and terminology codes in a vector database; s6, constructing a retriever, and retrieving related information from the vector database; and S7, combining the retrieved related parallel corpora with the to-be-translated content in combination with the footstone model to realize high-quality translation generation. According to the method, the flexibility and practicability of the system are greatly improved, the method is suitable for various application scenes such as received text processing and manuscript translation, cross-language communication is more convenient and efficient, and the method has extremely high practical value and wide application prospects.

The disclosure is directed to various ways of improving the functioning of computer systems, information networks, data stores, search engine systems and methods, and other advantages. Among other things, provided herein are methods, systems, components, processes, modules, blocks, circuits, sub-systems, articles, and other elements (collectively referred to in some cases as the “platform” or the “system”) that collectively enable, in one or more datastores (e.g., where each datastore may include one or more databases) and systems, the creation, development, maintenance, and use of a set of custom objects for use in a wide range of activities, including sales activities, marketing activities, service activities, content development activities, and others, as well as improved methods and systems for sales, marketing and services that make use of such entity resolution systems and methods as well as custom objects.

The invention discloses a title review material authenticity automatic verification method, system and platform based on deep network search and a large model. The method comprises the following steps: respectively generating and acquiring first data and second data corresponding to a to-be-processed material, and generating corresponding third data; creating a first model corresponding to the natural language, and generating fourth data corresponding to the search engine on the basis of the first model and in combination with the third data; based on the fourth data and in combination with a white list data source, performing search processing to generate fifth data corresponding to the to-be-processed material; constructing an auditing prompt word corresponding to the first data, and generating sixth data corresponding to the to-be-processed material according to the auditing prompt word in combination with a first model; wherein the sixth data are authenticity verification result data and reasoning basis data, and the system and the platform corresponding to the method can solve the problems of low manual auditing efficiency, insufficient retrieval automation degree, poor verification accuracy caused by semantic model closure and the like.