45results about "Web data indexing" patented technology

A computer-implemented method and system for determining search results for a search query for hypertext documents. The hypertext documents are reviewed to determine expert documents. When a query is received, the expert documents are ranked in accordance with the query. Then the target documents of the ranked expert documents are ranked to determine the search result set.

A programmable search engine system is programmable by a variety of different entities, such as client devices and vertical content sites to customize search results for users. Context files store instructions for controlling the operations of the programmable search engine. The context files are processed by various context processors, which use the instructions therein to provide various pre-processing, post-processing, and search engine control operations. Spam related and biased contexts and search results are identified using offline and query time processing stages, and the context files from vertical content providers associated with such spam and biased context and results are excluded from processing on direct user queries.

A system and method for automated search engine optimization (SEO) are disclosed. The automated SEO may analyze search query logs, or a search log database to determine popular concepts/units which may be automatically utilized to optimize a site or page for search engine results. The site or page is edited based on the automated SEO. In particular, the search log database may provide a unit frequency list reflecting the popularity of various units. The more popular units that are related to the content of the page may be used in the automated SEO of the page. The unit frequency list may be compared with the existing units of a page, so that the more popular concepts within the page may be emphasized.

The present invention provides systems and methods for automatically identifying name-like-strings in digital resources, matching these name-like-string against a set of names held in an expertly curated database, and for those name-like-strings found in said database, enhancing the content by associating additional matter with the name, wherein said matter includes information about the names that is held within said database and pointers to other digital resources which include the same name and it synonyms.

A method of managing on-line user data in a table-level distributed database system is disclosed. At a loading interface, a load request from a client device specifying a new or updated data table, which has been populated in a distributed file system, is received. The load request also specifies a location in the distributed file system and a schema of the data table. The loading interface creates new records in a metadata store, and each new record specifies a mapping between the data table and a particular instance that is to load such data table. Each record also specifies the location and schema for such data table. Based on the new records, each instance also loads the data table. Each instance publishes, in the metadata store, load status specifying that loading of the data table is complete so that such data table can be queried by a query client.

A system for prioritizing search results retrieved is described. One embodiment includes an inference, classification, and indexing subsystem configured to assign a local ranking to each occurrence of each data artifact in a collection of data artifacts obtained from on-line data objects, the local ranking assigned to each occurrence of each data artifact indicating a level of importance of that data artifact compared to other data artifacts obtained from the same on-line data object, the collection of data artifacts being indexed and organized by subject in at least one data structure, all data artifacts associated with a non-unique subject being associated with a single subject entry in the at least one data structure; and a search subsystem configured to assign, in response to the computerized search query, a global ranking to each data artifact in a set of data artifacts retrieved as search results from the collection of data artifacts, the global ranking of each data artifact in the set of data artifacts indicating a level of importance of that data artifact compared to the other data artifacts of like kind in the set of data artifacts, the global ranking of each data artifact in the set of data artifacts being based at least in part on the local rankings of the occurrences of that data artifact; prioritize the search results in accordance with the global rankings of the data artifacts in the set of data artifacts, the data artifacts of a given kind being grouped and arranged in descending order of global ranking; and present at least a portion of the prioritized search results to a user.

The embodiment of the invention discloses a method for identifying sensitive information, wherein the method includes steps of receiving target information, and extracting text information in the target information; calculating Hash value of the text information, and segmenting the text information and acquiring the word segmentation set when the Hash value of the text information is different from that of the preset feature sensitive information; calculating the Hash value of the segmented word in the word segmentation set; according to the Hash value of the segmented word in the word segmentation set, generating the similarity between the target information and the preset feature sensitive information; according to the similarity and/or the sematic analysis on the text information, judging that the target information is sensitive information. The invention further correspondingly discloses a device for identifying the sensitive information. The method and device for identifying sensitive information have high identifying accuracy in judging if the content published by a user is sensitive information.

The invention relates to a method for detecting suspected online water armies for microblog, and belongs to the technical field of computer application. The method for detecting suspected online water armies for the microblog comprises the following six steps: correspondingly related microblog data collecting; data preprocessing; user characteristics extracting; training set building; water armies detection model training; and unlabeled data predicting and distinguishing. Compared with the prior art, the method for detecting suspected online water armies for the microblog has the advantages that the full utilization of the data is realized, the group discovery is conducted conveniently and quickly, a complicated classification detection model is not needed to be established, so that the complexity of an algorithm is reduced, the modularity of the algorithm is relatively high, the algorithm can be put into large-scale data for calculation, and the stability is relatively high; a single user can be conducted on online water armies detection, a group of users in a certain event can be identified, the modularity of the method is extremely high and the method can be stably applied to large-scale data calculation framework.

In an information searching method, information content records are provided in a computer network, each information content record including a tag indicator forming a subset less than a whole of a context hashtag record. When a search query including at least one search term is received from a user, a search is performed of an index of context hashtag records for the at least one search term. The index of context hashtag records is based on data other than only the subset less than the whole of the context hashtag record. Based on the searching, a list of context hashtag record matches is generated, and a search result is returned to the user. The search result includes a list of information content records, each information content record on the result list being associated with at least one context hashtag record on the list of context hashtag record matches.

The invention discloses a picture identification method and a browser client. The picture identification method includes that receiving a first trigger operation for identifying a picture, wherein the picture is a picture in the current loading web page of the browser or a local picture obtained through the browser; sending the attribute information of the picture and/or picture to a server side; receiving and displaying keywords corresponding to the picture returned from the server side, wherein the keywords are found according to the attribute information corresponding to the picture and/or picture through the server side. By means of the picture identification method and the browser client, a user can identify the interested picture through the browser and obtain the keywords corresponding to the picture so as to obtain the other relevant information besides the picture appearance.

Methods of identifying web documents as relating to a market domain that would ordinarily be considered unrelated are presented. Market domain criteria can be defined that provide for classifying web documents as being related to the domain. The documents classified as related to the market domain form a training sample of documents used to establish correlations among brand term combinations found within the documents. If correlations are established among the terms in a combination, the term combinations can be assigned a similarity score indicating how similar the terms are considered to be. The term combinations can be used to search for additional web documents that could pertain to the market domain but would otherwise fail to satisfy the market domain criteria. The search results can be presented via a computer interface according to similarity scores.

The invention provides an associated information recommendation method and device. The method comprises the steps of obtaining m first weight values corresponding to m second-level information nodes associated with a first-level information node, wherein the m first weight values are obtained through click jump behaviors of a user, the first-level information node is an information node which is used by the user at current, and the m is a positive integer; determining associated information to be recommended to the user according to the m first weight values and preset n recommended information nodes, wherein the n is a positive integer. The associated information recommendation method and device have the advantages that since the click jump behaviors of the user are referred through the m first weight values, the problem of content focusing for recommending node information to the user through the contents of information nodes, uploaders or writers of the information nodes, types of the information nodes and the like is avoided; since the m first weight values are obtained by referring to the click jump behaviors of the user, the effect of recommending associated information to users through a big data concept is realized.

The invention provides a time series based topic development clustering analysis system and method. The method comprises: collecting topics to form a cumulative reading quantity time series; performing differential processing on the time series to obtain a topic hot degree time series; determining whether the topic is in a recession period, continuing to collect topics if not; calculating S-Euc and S-DTW between the topics if so, and clustering all topics. The system comprises: a data collection unit; a data processing unit that performs forward differential processing on the cumulative reading quantity time series to determine whether the topic is in the recession period, stores the topics that are not in the recession period in a first topic storage library, and stores the topics that are in the recession period in a second topic repository; a time series distance calculation unit that calculates the S-Euc and the S-DTW between the topics; and a topic clustering unit that clusters all the topics. The above system and method have high precision and a good clustering effect.

In a data mining system, data is gathered into a data store using, e.g., a Web crawler. The data is classified into entities. Data miners use rules to process the entities and append respective keys to the entities representing characteristics of the entities as derived from rules embodied in the miners. With these keys, characteristics of entities as defined by disparate expert authors of the data miners are identified for use in responding to complex data requests from customers.

The invention provides a graded and classified data search method and system. The method includes the steps that multiple search menus in public platform menus and Wechat Business platform menus are acquired, corresponding search menu options are acquired through click selection, and a first-grade search interface is started; the first-grade search interface displays multiple keywords extracted after industry data, region data and Wechat Business data are processed in a classified mode; merchant data and the Wechat Business data are classified, merchant name keywords and Wechat Business name keywords are extracted, multiple identification code pictures are correspondingly generated, keywords are acquired through click selection in the first-grade search interface, a second-grade search interface is started, merchant name keywords or Wechat Business name keywords in the second-grade search interface are acquired in a click selection mode, and corresponding identification code pictures are acquired; the identification code pictures are identified, and corresponding merchant data and the Wechat Business data are obtained. According to the graded and classified data search method and system, an accurate search result is obtained by identifying identification codes corresponding to the keywords extracted after data grading and classification.

The invention discloses a multi-granularity viewpoint mining method based on personal computer e-commerce comments, and the method comprises the steps: firstly obtaining the e-commerce comments from awebpage, and carrying out the automatic marking of the comments; preprocessing the e-commerce comments; recognizing 'aspect-viewpoint' pairs of single sentences in the preprocessed corpus, and distinguishing viewpoint words from aspect words; carrying out phrase-level sentiment classification on the context phrases of the aspect; and finally, carrying out text-level sentiment classification on the whole comments, so as to obtain viewpoint mining results of the whole commodity and all aspects. The problem that in the prior art, feedback information of all attribute aspects contained in comments cannot be automatically mined is solved.

This invention discloses a method and system for automatically dividing user levels, and relates to the technical field of data mining. The method comprises the steps of: S1, selecting original sample data; S2, selecting at least one type of user characteristic as a dimension for calculating distance; S3, determining number K of classifications; S4, selecting K users from the original sample data as initial classification centers; S5, measuring distance from the rest of each user in the original sample data to each of the current classification center; classifying the rest of each user into a nearest classification to finish the division of K classifications; S6, recalculating the classification center of each classification; S7, iterating the S5 and the S6; and stopping the iterative operation till a new classification center is equal to the original classification center or till the variable quantity is less than a specified threshold, so that the currently divided K classifications are the user level classifications required to be divided. The method can automatically divide the user levels, is precise and efficient, and can save the labor cost.

The invention relates to an advertisement data processing method and device, a computer device and a computer readable storage medium. The method comprises the following steps of capturing the advertisement data on a predetermined media by using a crawler, wherein the advertisement data comprises a group of advertisement information of a predetermined information type; extracting game type advertisement data from the advertisement data, and storing the game type advertisement data into a database; according to the service access demand of the first terminal, extracting advertisement information of an information type matched with the service access demand in game type advertisement data stored in a database; and processing the extracted advertisement information, generating the business data of a business type corresponding to the business access demand, and storing the business data into a database. According to the method, the service data obtained by analyzing and processing the game type advertisement data of different media are stored in the database, a user can conveniently access and obtain the needed related information of the game type advertisement data through the firstterminal, and the obtaining efficiency of the advertisement data is improved.

The invention discloses a method for detecting microblog emergencies. The method comprises the following steps: 1, carrying out time division on microblog data subjected to noise removal and word segmentation processing to obtain a microblog data set corresponding to a time window; 2, calculating a burst word judgment weight Wk (w) of each word w in all the microblogs contained in each time window, and taking words of which the Wk (w) is greater than a burst word threshold value as burst words, thereby obtaining a burst word set of each time window; and 3, based on the coupling degree of everytwo burst words in the burst word set, clustering the burst words in the burst word set by utilizing a clustering algorithm, and detecting an emergency of the time window based on a clustering result. Compared with an existing method, the microblog emergency detection method based on the emergency feature words has the advantages that the accuracy and the F value are greatly improved, that is, the detection result is more accurate.

The invention provides a multilayer retrieval method based on word segmentation. The method comprises the steps of obtaining retrieval basic split words of a retrieval text; combining the retrieval basic split words to form retrieval combined split words of the retrieval text; using the retrieval basic split words and the retrieval combined split words of the retrieval text to search for text datain a corpus to obtain a retrieval result of the retrieval text.

The invention discloses an authentication method, an authentication system, a storage medium and a big data authentication platform. The authentication method comprises the steps of: receiving an access request of a user; verifying the validity of the access request; and after the verification is passed, carrying out credit authentication on the user, and if the credit authentication is passed, allowing the user to access the big data platform. By verifying validity and creditability of the access request of the user, public opinion risk factors closely associated with an enterprise dominant are introduced in real time, and a credit limit threshold value of the enterprise dominant is dynamically adjusted, so that a timely and effective evaluation basis is provided for an evaluation subject, and the authentication method has the advantages of being real-time, efficient and reliable.

The invention provides an optimization method and system based on matching real-time performance in a publish-subscribe system, and the method comprises the steps: calculating the matching probabilityof subscriptions and events, carrying out differential treatment of subscriptions according to the matching probability, and forming subscription grouping types; and establishing a structure index ofthe data structure based on the subscription grouping category, and adjusting the subscribed grouping category and the position of the subscribed grouping category in the structure index in real timeaccording to the subscription matching probability change so as to improve the matching efficiency. Subscriptions with high matching probabilities are firstly processed in the matching process, so that subscriptions matched with events are determined earlier, and the real-time performance of event distribution is improved. A concise classification scheme is defined to group the subscriptions according to the matching probability of the subscriptions and the events. A lightweight subscription dynamic adjustment mechanism is established. An effective greedy algorithm is provided to solve the adjustment scheme. The high efficiency of a subscription classification and structure layering method (SCSL) is ensured, the efficiency is optimized and the invention can be configured to meet differentapplication requirements.

The invention discloses a search method and device of a distributed log. The method comprises the steps that a mode of an index filter selected by a user is determined; a screening condition which is based on the mode and inputted by the user is received; the index filter of the mode is operated, the index filter searches logs collected from a distributed network according to the screening condition, and the eligible log is obtained. The search device can conduct searching based on a saved original log file and prevents log information from being discarded in the saving process, and the searched log can be directly read by a technical staff, which helps the technical staff to rapidly make judgments according to the searched result.