Abstract: In a document classification method being executed by a processor of an electronic device, documents to be classified and categorical descriptions of the documents are received. Each document is classified into one or more categories according to a similarity between the document and each category specified in the categorical descriptions. The one or more categories into which each document is classified are outputted to an output device.

Abstract: In a web search optimization method, a keyword is inputted by a user, and an image of the user is captured to identify facial feature data of the user. When there is facial feature data matched with the identified facial feature data in a storage device of the electronic device, reference parameters which corresponds to the identified facial feature data are obtained, and web pages in a searched result relating to the keyword are ranked according to the reference parameters.

Abstract: A method for performing file search using a control computer receives an initial query expression from a client computer, divides the initial query expression into one or more keywords, determines related terms corresponding to each keyword, merges the keywords and the related terms to obtain an updated query expression, and generates search results according to the updated query expression. The method further calculates relevance scores among a plurality of documents in the search results and the updated query expression, ranks the documents in the search results according to the relevance scores, and displays the documents in the search results with the ranked order.

Abstract: A web searching method searches web pages by reference to a picture obtained from a search engine currently inputted by a user. Basic features of the reference picture are analyzed, and similarities between the reference picture and the pictures in a picture database is computed. Pictures from the picture database are selected according to similarities found, then web pages which contain the selected pictures are retrieved to find locations of the selected pictures in the web pages and obtain textual content around the selected pictures. Weightings of words and phrases in the textual content in each of the web pages are computed to select one or more of the words and phrases. The selected words and phrases are inputted into the search engine and the search result is displayed to the user.

Abstract: A web searching method isolates and presents relevant web pages based on analysis of keyword searches by a user and by users who have previously searched. First web pages are selected from a search result according to a keyword inputted by a user. Phrases in the first web pages are identified, and weightings of the phrases are computed. Phrases having higher weightings are selected. Related users previously requesting web pages related to the keyword are obtained and selected. Second web pages which were actually browsed by selected related users are obtained from the search result of the keyword, and phrases in the second web pages identified. Phrase intersections between the phrases from the second web pages and selected phrases from the first web pages are computed to realize an evaluation value of the selected related user. A help page showing searching histories of the related users is also displayed.

Abstract: In a method for creating data indexes for big data of a computing device, data lists are obtained from a data pool in a storage device, and a priority is set for each of the data lists. Data queues are created in the storage device, and the data lists are assigned to the data queues according to the set priorities. A node index is created for each data list stored in each of the data queues, and the data lists are deleted from the data queue after the node indexes creation. The method obtains a data list having a highest priority from the data pool if such a data list needs to be processed first, combines the node indexes to generate a root index for the data pool, and stores the root index of the data pool and the node indexes of the data lists in the storage device.

Abstract: In a method for displaying search results, a first query word is received from an electronic device. The method obtains a first group of web pages corresponding to the first query word, determines a degree of requirement of each web page in the first group for a user according to browsing records of the web pages in the first group by the user, and marks the degree of requirement of each of the web pages in the first group according to a preset rule.

Abstract: The present disclosure provides a method for searching related terms using an electronic device. The method adds time stamps to one or more electronic documents, and obtains related terms by calculating a relevance score between every two terms of the electronic documents within a specified time range. The method further calculates a time gap between each related term and a preset query term, obtains updated related terms that have a time relationship with the preset query term by removing specified related terms whose time gap is greater than a preset value, and obtains search results from a data source by performing a search operation according to the updated related terms.

Abstract: A method for searching related terms first calculates a direct relationship between every two of a plurality of query terms to obtain a direct related matrix, and calculates a related score between every two of the query terms to obtain a related score matrix. The method further calculates an indirect relationship between every two of the query terms according to the direct relationship and the related score, and determines indirect terms of each query term according to the indirect relationship between every two of the query terms.

Abstract: In a method for searching for related terms, one or more query terms are received from a client computer. The method determines hyponym terms of each query term, merges all the hyponym terms of the query terms into a set of the hyponym terms, and calculates a weight factor of each hyponym term in the set of the hyponym terms. The method further determines a specified number of hyponym terms according to the weight factor of each hyponym term, adds the determined hyponym terms into related terms of the query terms, and performs a search operation based on the hyponym terms of the query terms, the related terms of the query terms, and the query terms.

Abstract: The present disclosure provides a method for searching related terms using an electronic device. The method adds time stamps to one or more electronic documents, and obtains related terms by calculating a relevance score between every two terms of the electronic documents within a specified time range. The method further calculates a time gap between each related term and a preset query term, obtains updated related terms that have a time relationship with the preset query term by removing specified related terms whose time gap is greater than a preset value, and obtains search results from a data source by performing a search operation according to the updated related terms.

Abstract: A method for searching related terms first calculates a direct relationship between every two of a plurality of query terms to obtain a direct related matrix, and calculates a related score between every two of the query terms to obtain a related score matrix. The method further calculates an indirect relationship between every two of the query terms according to the direct relationship and the related score, and determines indirect terms of each query term according to the indirect relationship between every two of the query terms.

Abstract: A method searches related terms using an electronic device. The method adds time stamps to one or more electronic documents, and obtains related terms according to the electronic documents. The method further calculates a time gap between each related term and a preset query term, obtains updated related terms by removing specified related terms whose time gap is greater than a preset value, and obtains search results from a data source by performing a search operation according to the updated related terms.

Abstract: A method for searching related terms first calculates a direct relationship between every two of a plurality of query terms to obtain a direct related matrix, and calculates a related score between every two of the query terms to obtain a related score matrix. The method further calculates an indirect relationship between every two of the query terms according to the direct relationship and the related score, and determines indirect terms of each query term according to the indirect relationship between every two of the query terms.

Abstract: A method for performing file search using a control computer receives an initial query expression from a client computer, divides the initial query expression into one or more keywords, determines related terms corresponding to each keyword, merges the keywords and the related terms to obtain an updated query expression, and generates search results according to the updated query expression. The method further calculates relevance scores among a plurality of documents in the search results and the updated query expression, ranks the documents in the search results according to the relevance scores, and displays the documents in the search results with the ranked order.