Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

5 results about "Internet search engines" patented technology

Internet search engines are special sites on the Web that are designed to help people find information stored on other sites. There are differences in the ways various search engines work, but they all perform three basic tasks: They search the Internet -- or select pieces of the Internet -- based on important words.

System and method for determining optimized food combinations

A computer implemented method for use in conjunction with a computing device, system, network, and cloud with touch screen two dimension display or augmented / mixed reality three dimension display comprising: obtaining, analyzing and detecting user blood, saliva, hair, urine, stool, fingernail, height, weight and skin sampling analysis chemistry data, mapping the blood, saliva, hair, urine, stool, fingernail, height, weight and skin data into a database associated with a specific user, applying the data with optimization equations, mapping equations to food and beverage chemistry, scoring or ranking a plurality of optimized results such that a user may order food and beverage from a food / beverage distribution point or have food / beverage delivered to the user which has been specifically optimized for their specific biochemistry characteristic target ranges. The method is particularly useful in enhancing online internet search engine results.
Owner:CIRCLESX LLC

Multi-language law retrieval system

The invention provides a multi-language law retrieval system. The system comprises three modules: an input understanding module, a sub-query execution module and a networking search module, the input understanding module is used for receiving questions input by a user, understanding the specific questions input by the user by using a large language model, and finally decomposing user query into sub-query-national data pairs to form a sub-query list; the sub-query execution module receives a query list of the input understanding module, constructs a multi-language legal field retrieval system, realizes fusion of vector retrieval and BM25 through an Elasticsearch query function, and provides related sub-law article sets for countries and questions in law article data; the networking search module calls a search interface after understanding a user query, and searches related information on the Internet through an Internet search engine to realize knowledge expansion and information completion; and finally, jointly outputting the related sub-law article set and the related information on the Internet as a search result. Therefore, the retrieval precision of legal retrieval is improved.
Owner:RENMIN UNIVERSITY OF CHINA

Trusted retrieval enhancement method for external-related legal large model

The invention provides a credible retrieval enhancement method for an external-related legal large model. The multi-language query module receives question query input by a user, understands related countries input by the user, understands specific questions and establishes mapping of a country law domain and the questions; the multi-law domain legal resource recall module constructs a multi-language legal domain retrieval system, and retrieves a law article set related to the country according to the mapping of the national law domain and the problem; the external information networking retrieval module retrieves internet related knowledge through an internet search engine and supplements a static knowledge base; the judicial retrieval enhancement generation module inputs a law article set related to the country and internet related knowledge to obtain an answer under a judicial system corresponding to the country; and the multi-dimensional result aggregation generation module generates high-credibility answers through an external-related law big model. Through a query understanding mechanism, accurate disassembly and domain mapping of cross-country legal issues are realized, and the effect of efficiently processing multi-method-domain associated query which cannot be realized by an existing model is realized.
Owner:RENMIN UNIVERSITY OF CHINA +1

Practical training scene generation method based on retrieval enhancement and intelligent agent and related equipment

The invention provides a practical training scene generation method based on retrieval enhancement and an intelligent agent and related equipment. The method comprises the following steps: retrieving related information from a local vector database and an external internet search engine according to standardized query after semantic rewriting by adopting a'local + external 'dual-channel knowledge fusion mechanism; in a local channel, performing multi-path recall by combining keyword matching and dense vector semantic retrieval, dynamically allocating weights according to the inquired terminology density, and performing preliminary sorting on two paths of results through a reciprocal sorting fusion algorithm; then the fused local retrieval result is sent to a cross encoder for fine-grained correlation reordering, and an optimal context is screened out; and finally, injecting the context and the rewritten standardized query into a preset prompt template together, generating a code script by a large language model, and generating a network practical training scene according to the script. The automation process reduces the practice threshold, and improves the teaching and learning efficiency.
Owner:BEIJING UNIV OF POSTS & TELECOMM

Categorical search using visual cues and heuristics

A graphical user interface with visual categories having a logical relationship for defining search queries for internet search engines or other search applications and devices. Users can input search words, phrases, or other data that apply to the interface categories. A visual metaphor for categories of logical operators, search operators, and search parameters assists users and facilitates generating syntactically optimized search queries. There is also the capability for users to use the interface to specify the relative importance and proximity of words, phrases, or other data. Searches can be filtered, focused, iterative, and persistent.
Owner:ROSS MICHAEL M