Abstract: A method, device and medium for constructing a knowledge base is described, wherein the knowledge base construction based on retrieving text information, determining whether the text information includes at least first information according to a predefined property description pattern, extracting the first information from the text information when the first information is determined to be included in the text information, and constructing the knowledge base based on the first information and an entity corresponding to the first information.

Abstract: Methods and devices for mining an information template are provided. A method may include forming a modeling information set comprising a plurality of modeling information items. The method may further include creating a plurality of encrypted information items by encrypting respective numerical information items included in the plurality of modeling information items. The method may further include clustering the plurality of encrypted information items to create at least one information template. According to the present disclosure, an information template may be mined through analysis of a plurality of modeling information items, and numerical information items included in the modeling information items may be encrypted during the template mining process, which may prevent users' private information from being disclosed by the mined template, so that a more secure method for mining an information template may be provided.

Abstract: An alarm method is disclosed. The method may comprise acquiring a monitoring video, detecting whether the monitoring video contains a moving target, determining whether the moving target is a preset object if it is detected that the monitoring video contains the moving target, and forbidding sending corresponding alarm information to a terminal if the moving target is determined to be the preset object.

Abstract: A method and device for determining semantic similarity between two character strings are disclosed. The two character strings are segmented into sequences of words or phrases which represent the correlation between the characters. Edit distance from the first sequence to the second sequence is calculated based on a predetermined algorithm. A minimum semantic distance is then determined from the edit distance by considering the word/phrase pairs appearing in both sequences and the relationship between the cost of the various operations performed to convert the first sequence into the second sequence. The semantic similarity between the two character strings is then determined and normalized from the minimum semantic distance.

Abstract: A method, device, and a non-transitory computer-readable storage medium are provided for identifying information in an electronic message. The information from the electronic message may be identified according to a template that has been trained to classify the information included in the electronic message.

Abstract: The present disclosure is related to a digital image processing method and apparatus for generating a description of contents depicted in a digital image based on at least one or more of identity information of a person depicted in the digital image and digital image information.

Abstract: A facial recognition solution is disclosed that includes adding a specified numeric pixels to an edge area of a digital image to acquire an enhanced digital image and then performing a facial recognition process on the enhanced digital image to determine a human face from the digital image.

Abstract: Described are an alarming method and device, belonging to the technical field of Internet. The method includes: acquiring a monitoring video from a photographing device, wherein the photographing device captures the monitoring video by monitoring a region including a sensitive area; judging whether a monitoring target exists in the sensitive area by analyzing the monitoring video; and when the monitoring target exists in the sensitive area, sending alarming information to a terminal device to enable the terminal device to prompt alarm.

Abstract: A method for fingerprint recognition utilizes an auto encode decode network to perform feature extraction on a first fingerprint image acquired by a fingerprint sensor and a second fingerprint image retrieved from a database. First and second fingerprint features corresponding to the first and second fingerprint images are thus obtained. The first and second fingerprint features may have equal dimensionality. Dimensionality reduction may be performed on the first and second fingerprint features to respectively obtain third and fourth fingerprint features. The third and fourth fingerprint features may have equal dimensionality. A cosine distance between the third and fourth finger print features may determine whether the first fingerprint image and the second fingerprint image belong to a same fingerprint.

Abstract: A method for conducting classification model training includes: acquiring sample feature vectors of a plurality of users according to at least one feature set for each of the users, wherein the at least one feature set for a user is determined based on at least one sample message of the user; determining gender identifiers of the users according to the sample feature vectors; and conducting training based on the sample feature vectors and the gender identifiers corresponding to the sample feature vectors to construct a gender classification model.

Abstract: The present disclosure relates to a method, device and computer-readable storage medium for video display. The method includes detecting a first trigger action on a first screen area; determining a trigger position corresponding to the first trigger action; determining a first set of parameters of an observation window based on the trigger position, wherein the observation window is centered on the trigger position; generating a second screen area based on the first set of parameters of the observation window; enlarging the second screen area to a size of the first screen area; and displaying the enlarged second screen area in place of the first screen area.

Abstract: A method and device for determining semantic similarity between two character strings are disclosed. The two character strings are segmented into sequences of words or phrases which represent the correlation between the characters. Edit distance from the first sequence to the second sequence is calculated based on a predetermined algorithm. A minimum semantic distance is then determined from the edit distance by considering the word/phrase pairs appearing in both sequences and the relationship between the cost of the various operations performed to convert the first sequence into the second sequence. The semantic similarity between the two character strings is then determined and normalized from the minimum semantic distance.

Abstract: The disclosure relates to a method, device, and non-transitory computer-readable storage medium for recommending reference documents in real-time. Aspects of the disclosure provide a method for recommending reference documents. The method includes obtaining content information from a target document that is being accessed by a user, obtaining key information from the content information by parsing the content information according to a parsing rule, extracting search words from the key information, performing a search based on the search words, obtaining reference documents including the search words based on the search, and displaying the reference documents.

Abstract: Method and apparatus are disclosed for organizing photos into albums. Faces in the photos may be recognized and classified as faces of interest or faces of irrelevance. Electronic photo albums may then be established where each electronic album corresponds to a unique face among the faces of interest. Each photo may be assigned to one or more electronic album according to the faces of interest contained in the photo.

Abstract: The present disclosure is related to a digital image processing method and apparatus for generating a description of contents depicted in a digital image based on at least one or more of identity information of a person depicted in the digital image and digital image information.

Abstract: A method, device, and a non-transitory computer-readable storage medium are provided for identifying information in an electronic message. The information from the electronic message may be identified according to a template that has been trained to classify the information included in the electronic message.

Abstract: Methods and devices for mining an information template are provided. A method may include forming a modeling information set comprising a plurality of modeling information items. The method may further include creating a plurality of encrypted information items by encrypting respective numerical information items included in the plurality of modeling information items. The method may further include clustering the plurality of encrypted information items to create at least one information template. According to the present disclosure, an information template may be mined through analysis of a plurality of modeling information items, and numerical information items included in the modeling information items may be encrypted during the template mining process, which may prevent users' private information from being disclosed by the mined template, so that a more secure method for mining an information template may be provided.

Abstract: An image recognition method and device for implementing game play features of a game related to the field of image processing is disclosed. By allowing a communication device controlled by a player participating in the game to capture images and apply the image recognition features on the captured images, game play operations may be applied to other players included in the captured image within a context of the game play of the game.

Abstract: The present disclosure is directed to an information processing method, an information processing apparatus, a terminal and a server. One method of the present disclosure involves receiving one or more data templates from a remote information processing server and storing the data templates in a local database, receiving a data object from a third-party server, and when the data object has been received, determining whether the data object comprises a particular type of data, comparing a determined data type to the one or more data templates in the local database to identify a matching data template, extracting data elements from the data object using the matching data template, and utilizing the extracted data elements to automatically perform a task. The embodiments of the present disclosure may automatically identify and extract information from data objects in an accurate and efficient manner, and enable users to take certain actions using the extracted information, thereby improving a user's experience.

Abstract: An alarm method is disclosed. The method may comprise acquiring a monitoring video, detecting whether the monitoring video contains a moving target, determining whether the moving target is a preset object if it is detected that the monitoring video contains the moving target, and forbidding sending corresponding alarm information to a terminal if the moving target is determined to be the preset object.