550results about How to "Improve click-through rate" patented technology

According to one aspect of the present invention a method and apparatus are described for improving advertising conversions on the Internet. An analysis engine is provided that analyzes raw advertising metrics in order to identify improvements. A treemap based visualization engine allows the user to visualize a tree in two dimensional space. In one embodiment, an action engine includes rapid one-box recommendation that allows the user to take an action to improve an advertising campaign. According to another aspect, a system and method for managing a plurality of advertising networks is also provided. A typical embodiment of the management system integrates the analysis engine, visualization engine, and action engine in order to optimize a user/manager's time and effort in organizing, improving, and managing advertising campaigns across a plurality of advertising networks. The presentation and organization (rendered by the visualization engine) of visual displays of advertising information (compiled by analysis engine) effectively reduces the workload in managing the plurality of advertising networks, additionally, recommendations can be based on advertising information (supplied by the action engine). In one example, visual cues in the form of color designations, bring the user's attention to advertising nodes on which actions are estimated to have the significant impact. The definition of what a significant impact is may be established by default or in other embodiments can be configurable by each particular user or manager. Once the user's attention is brought to a particular advertising node, actions and alerts can be recommended to improve an individual advertising campaign, ad group, keyword, ad copy, and/or ad. In one embodiment, by providing an interface to access to other networks with their own advertising campaigns a plurality of networks can be managed.

Behavior-based associations are extrapolated to items for which the quantity of collected user activity data is insufficient to create meaningful or reliable behavior-based associations (“behavior-deficient” items). The behavior-based associations are extrapolated based on content-based associations, or another type of “substitutability” association, between the behavior-deficient items and other items. The items can be any type of item for which user behaviors (e.g., purchases, accesses, downloads, etc.) can be monitored and analyzed to detect behavior-based associations, and for which item content or other available information can be used to assess item substitutability. For example, the items can be products represented in an electronic catalog, web pages or other documents accessible on a network, or web sites.

In a web page advertising method with a learning mechanism, a plurality of advertisement files are transformed to vector equations in advance. Content of a web page file that is being displayed is analyzed and transformed to a vector equation in real time. An advertisement search conditional set is determined according to the analyzed web page file. The advertisement files are subjected to a similarity analysis using vector computations so as to select at least one advertisement file that has a high correlation to the overall concept of the content of the web page. Adjustment and training with the advertisement file are conducted through analyzing actions of a user, thereby increasing considerably the click rate of the advertisement file and achieving full utilization of advertising resources. An advertising agent employing the web page advertising method purchases advertising spaces from a medium and bills advertisers on a pay-per-click basis.

There are provided methods and systems for inferring a user's interests from user-generated tags of web-based content. In accordance with the invention, user-generated tags from viewing web-based content are collected over a predetermined period of time. A subset of distinct or unique tags is identified from among the collected tags. A z-score is calculated for each identified distinct tag, where the z-score is a measure of the statistical significance of the tag. The subset of distinct tags is then sorted based on their corresponding z-score. All distinct tags having a corresponding z-score lower than a predetermined threshold are rejected and the remaining distinct tags, having a corresponding z-score higher than the threshold are used to infer a user's interest. The ability to infer a user's interests from the remaining distinct tags may thus benefit web-based applications by achieving a high degree of accuracy in predicting the interests of users by leveraging on the use of the user generated content tags and keywords.

The invention relates to a news recommendation method and system, and relates to the technical field of information. The method and system aim to push news to different types of users who are interested in the news. The method includes the steps of extracting features of search and query data, according to behaviors of a certain type of users on the search and query data, calculating and recording interest weights of the features of the search and query data on the type of users, extracting features of multiple pieces of news to be pushed, searching for the interest weights of the features of the news to be pushed from the recorded features and the interest weights, calculating interest scores of the news to be pushed for the type of users, and pushing the news to the type of users according to the interest scores of the news to be pushed for the type of users. By means of the method and system, interests of different types of users can be analyzed, and the news according with the interests of the users is pushed to the users.

The invention discloses a content recommendation method and device, electronic equipment and a storage medium. The method relates to natural language processing and video content understanding directions in the field of artificial intelligence, and a playing page of a target video can be displayed; when a video clip currently played by the target video belongs to a preset video clip, a recommendedcontent bullet screen corresponding to the preset video clip is displayed in a playing area of the target video; based on a trigger operation for description information in the recommended content bullet screen, recommended content corresponding to the description information is displayed; according to the embodiment of the invention, the recommended content bullet screen can be displayed on theplaying area of the target video watched by the user, and the recommended content corresponding to the recommended content bullet screen is associated with the video clip of the target video watchedby the user, so that the recommended content has high attraction to the user, the interest of the user can be increased, and the improvement of the interaction degree of the user and the click rate ofthe recommended content is facilitated.

The invention provides a method and device for showing a search result. The method comprises the steps that a current search term input by a user is received; user information is determined, a pre-established first model corresponding to the user is determined according to the user information, historical data of the user are recorded in the first model, and the historical data comprise historical search terms and corresponding historical selection results; the historical search term matched with the current search term is determined in the historical data, the historical selection result corresponding to the matched historical search term is determined, and an on-line recommendation result is determined according to the historical selection result; the search result is shown to the user, the search result comprises the on-line recommendation result or a result obtained after the on-line recommendation result is processed. By the adoption of the method for showing the search result, an individualized search service can be provided for the user, and therefore the search result can better meet the demands of the user.

The invention discloses a method for showing advertisements in a webpage and a system. The method comprises the following steps: receiving a webpage access request of a user, extracting user identification in the webpage access request; searching a preset user cluster group according to the user identification, wherein each user cluster is constructed according to access character information of multiple users and corresponds with a specific advertisement; and determining the user cluster to which the user belongs, and showing the specific advertisement corresponding to the user cluster for the user in the webpage. Therefore, the method realizes showing advertisement in the webpage by selecting the specific advertisement corresponding to the user cluster to which the user belongs, whereinthe user cluster is constructed by the access character information of multiple users to represent the access character of the users, so the selected specific advertisement is suitable for the user, and the probability for clicking the specific advertisement by suitable users is greatly improved.

The invention discloses a message pushing method and device, and relates to the technical field of data processing. The message pushing method and device mainly aim at solving the problems that usersaiming at the message pushing screening in different scenes in the prior art are random, or the pushing is performed through large-range message link sending, so that the pushed users have lower individuation; directionality is not realized; when the users do not click the pushed messages, the flow waste can also be caused; the click rate is lower; useless pushing is caused. The method provided bythe invention has the main technical scheme that the method comprises the following steps of obtaining user portraits of messages to be pushed; extracting attribute features in the user portraits; determining the message pushing users by combining the user screening prediction models, wherein the user screening prediction models are used for screening the users expected to have the clicking behaviors after the message pushing according to the attribute features; pushing the messages for the user. The method and the device are mainly used for message pushing.

This invention concerns optimal ad selection for Web pages by selecting and updating an attribute set, obtaining and updating an ad-attribute profile, and optimally choosing the next ad. The present invention associates a set of attributes with each customer. The attributes reflect the customers' interests and they incorporate the characteristics that impact ad selection. Similarly, the present invention associates with each ad an ad-attribute profile in order to calculate a customer's estimated ad selection probability and measure the uncertainty in that estimate. An ad selection algorithm optimally selects which ad to show based on the click probability estimates and the uncertainties regarding these estimates.

The invention provides an information recommendation method and device based on artificial intelligence, electronic equipment and a computer readable storage medium. The method comprises the steps ofobtaining a historical behavior information sequence of a user; according to the vector representation of each piece of information in the historical behavior information sequence, determining a behavior feature vector representing the behavior feature of the user; iteratively encoding the behavior feature vector for multiple times to obtain a user interest vector representing the interest of theuser; and determining a plurality of pieces of information satisfying similar conditions with the user interest vector to execute recommendation operation for the user. According to the invention, therecommendation accuracy and the click rate index of the recommendation information can be improved.

The embodiment of the invention provides a model generation method, a model generation device, a recommendation method and a recommendation device. The model generation method in the embodiment of the invention comprises the following steps: acquiring a document content feature vector of each entity in a mapping knowledge domain, logical relationship feature vectors among the entities, a user behavior relationship feature vector of each entity and at least one of feature vectors of each entity, then performing machine learning according to the document content feature vectors, the logical relationship feature vectors, the user behavior relationship feature vectors and the at least one of the feature vectors, and generating a deep fusion model. Therefore, according to the technical scheme provided by the embodiment of the invention, the deep fusion model can be generated by integrating various relationships among the entities and can be used for acquiring the surprise degree among the entities, so that the entities can be recommended to users based on the surprise degree, the search recommendation requirements of the users are met, and the click rate of the recommended entities is improved.

An embodiment of the invention provides an advertisement recommending method and an advertisement recommending server. The method comprises the steps as follows: obtaining webpage visit information and advertisement click information, wherein the webpage visit information is used for indicating n webpages visited by m users, and the advertisement click information is used for indicating x advertisements clicked by the m users on the n webpages; predicting click probabilities of the x advertisements when the i user in the m users visits the j webpage according to the webpage visit information and the advertisement click information; determining novelty factors corresponding to the x advertisements respectively; determining p advertisements to be recommended to the i user in the x advertisement according to the click probabilities of the x advertisements and the novelty factors corresponding to the x advertisements respectively. By means of the advertisement recommending method and the advertisement recommending server, the click rate of the advertisements can be increased, and the user experience can be improved.

The invention provides a control method for issuing advertisement information to a user terminal according to user ordering information in an advertisement issuing control system, comprising the following steps: 1, obtaining user ordering information; and 2, issuing advertisement information according to the user ordering information. The invention also provides a control device for issuing the advertisement information to the user terminal according to the user ordering information in the advertisement issuing control system, and the advertisement issuing control system for issuing the advertisement information to the user terminal according to the user ordering information. The invention permits a user to put forward the requests of advertisement contents that the user hopes to browse to the advertisement issuing control system, and issues the advertisement contents according to the requests, thereby issuing advertisements in which the user is interested and farthest avoiding that the user generates boredom emotions to the advertisements. The advertisement issuing control system applying the invention can enhance the clicking rate of the advertisements so as to enhance the effective rate of the main issuing advertisements of the advertisements, expand the influence of the products of the advertisements, bring more benefits for advertisement suppliers and realize the best effect of advertisement issuing.

Ranking product information is disclosed, including: determining, for each of a plurality of pieces of product information, a category grading value and a plurality of attribute grading values associated with that piece of product information; determining a plurality of user demand values corresponding to the plurality of pieces of product information based at least in part on the category grading value and at least one of the plurality of attribute grading values associated with each of the plurality of pieces of product information; and ranking the plurality of pieces of product information based at least in part on the corresponding plurality of user demand values.

The embodiment of the invention provides a search-word pushing method, device and terminal, relates to the field of data processing technology, and mainly aims to pertinently push search words for different users, and increase click rates of popular search words. The technical solution adopted by the embodiment of the invention includes: acquiring identity information, location information and current time information of a user; extracting user hot-search words, which correspond to the user, from user history behavior data, which correspond to the identity information, to obtain the first search word set; acquiring public hot-search words, which correspond to the location information and the time information, to obtain a second search-word set, wherein the public hot-search words are search words used by public users for executing search operations at a location and a time corresponding to the location information and the time information; carrying out sorting according to hot degreesof the search words, and respectively screening out a preset number of to-be-pushed search words from the first search-word set and the second search-word set; and pushing the to-be-pushed search words to the user. The method is mainly used for pushing the search words.

The invention discloses a method for extracting a text from a web page and a device thereof, and relates to the technical field of networks. The method and the device rapidly and accurately extract the text from the web page. The method comprises the following steps: obtaining a start tag and an end tag of a page segment; determining a start position and an end position of the page segment according to the start tag and the end tag; computing a weight value of the page segment; and extracting the page segment with maximum weight value from the web page to be taken as the text of the web page. The device comprises an acquisition module used for acquiring the start tag and the end tag of the page segment; a segmentation module used for determining the start position and the end position of the page segment according to the start tag and the end tag acquired by the acquisition module; a computation module used for computing the weight value of the page segment determined by the segmentation module; and an extraction module used for extracting the page segment with the maximum weight value computed by the computation module from the web page to be taken as the text of the web page. The technical proposal provided by the invention can be widely applied to network systems and devices for reproducing contents.

The invention relates to the technical field of video image processing, in particular to an image special effect processing method and device, a live video terminal, computer equipment and a computerstorage medium. The image special effect processing method comprises the steps of receiving a special effect processing instruction, and calling a special effect template for video image special effect processing according to the special effect processing instruction; wherein the special effect template is provided with a plurality of special effect processing channels; obtaining multiple paths ofimage data according to an input source path, and transmitting the corresponding image data to each special effect processing channel of the special effect template; and performing parallel special effect processing on each path of image data through the special effect processing channel, and merging each path of image data after special effect processing to obtain a target special effect image.According to the scheme, rich special effect design effects can be obtained, and the scheme is suitable for diversified requirements of current users on small videos and live broadcast contents.