30results about How to "According to the mind" patented technology

The invention belongs to the technical field of mobile terminals and particularly relates to a quick screen luminance adjustment method and a quick screen luminance adjustment device. The quick screen luminance adjustment method is applied to a mobile terminal without side bezels, and a side edge area and a screen luminance adjustment gesture are preset for the mobile terminal. The method includes: when the mobile terminal is in a mis-touch prevention mode, detecting a gesture in the side edge area; judging whether the detected gesture is the screen luminance adjustment gesture, and if yes, adjusting the screen luminance of the mobile terminal according to the gesture in the side edge area. Due to the fact that the condition of mis-touch prevention is added when whether a current operation is a luminance adjustment operation or not is judged, misoperation can be avoided while the mobile terminal is enabled to uniquely judge that the current operation of the user aims to realize luminance adjustment rather than something else, so that the user can adjust the mobile terminal more desirably.

The invention discloses a method and a device capable of shooting photographs. The method comprises the following steps: receiving a continuous shooting command and obtaining N photographs in a preset time, wherein N is an integer being greater than or equal to 2; respectively identifying a preset concerned area in each photograph; and synthesizing at least one photograph with the concerned area which most meets the preset condition in other N-1 photographs except a base photograph with the base photograph to obtain a final photograph, wherein the base photograph is any one of N photographs. Through the scheme of the embodiment, the final photograph obtained by synthesizing at least one photograph with the concerned area which most meets the preset condition in multiple photographs with the base photograph is high in quality and can more meet the intention of the user.

The invention discloses that the invention relates to the technical field of mobile terminals, in particular to a method and device for quickly adjusting screen brightness, and belongs to the technical field of mobile terminals. The method for quickly adjusting the brightness of the screen is applied to a mobile terminal without a side frame. The mobile terminal is preset with a side edge area and a gesture for adjusting the screen brightness. The method includes: when the mobile terminal enters the anti-mistouch mode, detecting Gestures that occur in the area; determine whether the detected gesture is a gesture to adjust screen brightness, and if so, adjust the screen brightness of the mobile terminal according to the gestures that occur in the side edge area. In the process of judging that the current operation is brightness adjustment, the present invention adds the condition of preventing false touches, which can not only avoid misoperations, but also enable the mobile terminal to uniquely determine that the current operation of the user is brightness adjustment, not others. This makes the user's adjustment to the mobile terminal more satisfying.

The invention discloses a method for recommending friends based on a network account. The method comprises the following steps: upon receiving a friend recommendation request, acquiring a first network account contained in the friend recommendation request; based on the preset filtering rules and the first network account, determining a network account set which meets the filtering rules; conducting separate matches between the key word set of the first network account and a key word set of each network account in the acquired network account set, determining a degree of matching of each network account in the network account set; recommending the network accounts having a degree of matching met the preset matching conditions in the network account set to the first network account. The invention also discloses an apparatus for recommending friends based on the network account. The apparatus can perform matching based on the key word set, recommends friends who have the same or similar interests, hobbies and habits as the user of the first network account based on the key words set, and has a flexible recommendation method and more accords with user's interests.

The present invention claims a song recommendation method based on a hybrid recommendation algorithm, which includes the following steps: the user requests to play a song, obtains the user's song play log, obtains the user-song scoring matrix, and determines it from the preset user attribute item set Go out with the target attribute item of described play song request, and utilize the user-song score matrix of establishment, carry out the preliminary recommendation of song; According to song play log, calculate the user preference degree of each song in the song play log; Determine recommendation list, and recommend the recommended list to the user; if half of the results are the same as the initial recommendation of the song, then click here to recommend, if not, select the song with the second highest score as the target song, and repeat the above steps to fill until it exceeds If half of them are the same, the song list will be recommended to the user. The hybrid recommendation method of the present invention can improve the accuracy of song recommendation.

The invention requests protection of a method for alleviating data sparsity in a recommendation system based on step-by-step dynamic filling. The method firstly preprocesses the user behavior data, and establishes the user-gourmet merchant scoring matrix; secondly, selects the target user to calculate the user similarity, and selects the user whose similarity is greater than the threshold α as the preselected similar neighbor user of the target user; then, calculates the user common score The difference mean value, select the pre-selected neighbor users whose common score difference mean value is less than the threshold β as the final similar neighbor users, and use the neighbor users to fill the scoring matrix in the first step; finally, for the remaining unfilled data in the scoring matrix, similar The degree threshold method and the mean value of the common rating difference are used to select similar merchants, and the second step is used to fill the scoring matrix with similar merchants. The present invention alleviates the sparsity problem of scoring matrix in the gourmet business recommendation system to a certain extent, and improves the accuracy of the recommendation system.

The present invention claims to protect a news recommendation method for solving data sparseness, which includes the following steps: receiving a voice request for broadcasting news, screening the user's historical behavior data of the voice request, and obtaining a news listening user-news broadcast scoring matrix , determine the target attribute item associated with the voice request from the preset user attribute item set based on the screening result; and construct a historical scoring record set for each news listening user and each news broadcast terminal, and construct a user set at the same time, according to Sorting the users in the user set from large to small according to the number of users’ rating news broadcast terminals; constructing the final similar user set, using the similar user set of the target user to fill the news listening user-news broadcast scoring matrix; the rest of the scoring matrix The unpopulated data is filled again by similar news broadcasting terminal, and the news broadcasting terminal makes a preliminary recommendation; the target news is screened out for recommendation. The invention can improve the accuracy of news recommendation.

The invention discloses a method for determining the weight of an answer, which includes: obtaining a behavior track; the behavior track represents a user's satisfaction degree to a target answer provided to the user; and determining the weight of the target answer according to the behavior track. The invention also discloses an answer weight determination device, a storage medium, an answer determination method, a device and a storage medium.

The present invention provides a training planning anchoring method, device, and middle platform based on fusional geographic relationships, including: receiving training planning data, the training planning data including three different fusion geographic relationships; acquiring previous historical planning data, based on the The historical planning data sorts three different fused geographic relationships to obtain the first geographic relationship, the second geographic relationship, and the third geographic relationship; based on the first geographic relationship and the second geographic relationship, at least two types in the database The first screening and the second screening are performed on the training grounds to obtain at least one first-type training ground and at least one second-type training ground; The consistency comparison of the type training sites is performed, and the first-type training sites and the second-type training sites whose consistency results are greater than the threshold are anchored to obtain a training plan.

The invention relates to a picture processing method and apparatus. The method includes the following steps that: each of N pictures obtained under a continuous capture mode is segmented into M sub-arrays, wherein N and M are integers greater than or equal to two; a sub-array which most satisfies a preset standard is selected from the i-th sub-array of the first picture to the i-th sub-array of the N-th picture as the i-th sub-array of a composite picture, wherein i is an integer from 1 to M; and the first sub-array to the M-th sub-array of the composite picture are re-combined so as to form a final picture. According to the picture processing method provided by the embodiments of the present invention, each of the N pictures obtained under the continuous capture mode is segmented into the plurality of sub-arrays; the sub-array which most satisfies the preset standard is selected from the N pictures; a new picture can be obtained through combination; and therefore, a high-quality picture can be obtained, the requirement of a user can be satisfied.

The present invention claims protection for a mixed recommendation method for an online shopping mall system, which includes the following steps: the user inputs purchase request information, and obtains the user's historical purchase information, screens the user's historical purchase information, and acquires the information related to the online shopping mall. and user-related information, get the user-online product rating matrix, construct a historical scoring record set, construct the final similar product set, use the similar product set of the target product to fill the user-online product rating matrix; and use similar online products to Fill in the user-online product rating matrix again to make preliminary recommendations for online products; divide the user's historical purchase data into single-dimensional user feature values ​​and multi-dimensional user feature values ​​to determine the recommendation list. The hybrid recommendation method of the present invention can improve the accuracy of commodity recommendation.

The invention discloses an answer weight determination method. The method comprises the steps of obtaining a behavior track; wherein the behavior track represents the satisfaction degree of the user on a target answer provided for the user; and determining the weight of the target answer according to the behavior track. The invention further discloses an answer weight determination device, a storage medium, an answer determination method and device and a storage medium.

The invention requests to protect a recommendation method of a TV program. The method comprises the following steps: firstly receiving language information input by a user; obtaining a characteristicelement according to the language information, retrieving and evaluating the electronic program information in an electronic program table database by taking the characteristic element as the retrieval keyword, and extracting the related electronic program information according to the similarity; constructing a statistical model by utilizing the characteristic set and the machine learning method;matching the programs in the electronic program table database by using the statistical model; outputting a matching result to the user; secondly, sending the program request information by the user,acquiring score information on the program by the user to obtain a user-program score matrix, and initially recommending the program; if the initially recommended program is half same as the matchingresult, recommending based on this situation, if the initially recommended program is different from the matching result, selecting the program with the second-highest score as the target program, andrepeating the above steps to perform the filling, and recommending the program list to the user until more than half recommended programs are same as the matching result. The accuracy of the programrecommendation can be improved through the mixed recommendation method.