Abstract:
A technology for allowing anyone to easily create interactive media capable of easily recognizing a user interaction by using a stored image is provided. A system according to the present invention includes an image reconstruction server, an image ontology, and an image repository. The image reconstruction server includes an image reconstruction controller, a natural language processing module, and an image search module. The image reconstruction controller of the image reconstruction server receives a scenario based on a natural language from a user and searches for images desired by the user by using the natural language processing module, the image search module, and the image repository. The natural language processing module of the image reconstruction server performs a morphological analysis and a syntax analysis on the scenario input by the user as a preliminary operation for the search of the image ontology. The image search module of the image reconstruction server automatically generates an ontology search query sentence, such as SPARQL, by using a result of natural language processing, and searches the image ontology by using the generated query sentence.
Abstract:
Provided are a clustering method using broadcast content and broadcast related data and a user terminal to perform the method, the clustering method including creating a story graph with respect to each of a plurality of scenes associated with broadcast content based on the broadcast content and broadcast related data, and creating a cluster of a scene based on the created story graph.
Abstract:
A crime type inference system based on text data, may include: a keywords dictionary construction unit configured to receive crime source data, and generate a crime type keywords dictionary by extracting crime keywords; a data set construction unit configured to generate a dataset for crime type learning by using the crime source data and the keywords dictionary; a crime type prediction model training unit configured to generate a crime type prediction model by using the dataset, and train the crime type prediction model; and a crime type inference unit configured to infer a crime type by using new crime data.
Abstract:
A method of improving performance of classification on the basis of mixed sampling is applied. The present invention is directed to providing a method and apparatus for improving the performance of classification on the basis of mixed sampling that are capable of, when learning a model that classifies types using deep learning by dividing the entire data set and using the divided data set for training, validation, and testing, applying different sampling techniques by types of data according to the characteristics of the training data in order to improve the classification performance.
Abstract:
The present invention provides a malicious application detecting apparatus based on a visualization similarity, including: a first storing unit which classifies malicious applications for every group in accordance with characteristics and stores the malicious applications; a second storing unit which stores a target application; an image generating unit which analyzes the malicious applications to generate first visualization images and analyzes the target application to generate a second visualization image; a representative image selecting unit which selects representative images for every group using a similarity of the first visualization images; and a determining unit which compares the representative images with the second visualization image to determine whether the target application is a malicious application.
Abstract:
The present invention relates to an emergency reporting system and method for the socially disadvantaged. The emergency reporting system for the socially disadvantaged according to the present invention includes a user terminal configured to receive an emergency report input in a preset manner according to environment information set by the socially disadvantaged, generate an emergency report message, and transmit the emergency report message, and a server configured to receive the emergency report message and transmit a dispatch notification signal.
Abstract:
Disclosed is an apparatus and method for tagging a topic to content. The apparatus may include an unstructured data-based topic generator configured to generate a topic model including an unstructured data-based topic based on content and unstructured data, a viewer group analyzer configured to analyze a characteristic of a viewer group including a viewer of the content based on a social network of the viewer and viewing situation information of the viewer, a multifaceted topic generator configured to generate a multifaceted topic based on the topic model and the characteristic of the viewer group, a content divider configured to divide the content into a plurality of scenes, and a tagger configured to tag the multifaceted topic to the scenes.
Abstract:
Disclosed is a method and apparatus for generating a title and a keyframe of a video. According to an embodiment of the present disclosure, the method includes: selecting a main subtitle by analyzing subtitles of the video; selecting the keyframe corresponding to the main subtitle; extracting content information of the keyframe by analyzing the keyframe; generating the title of the video using metadata of the video, the main subtitle, and the content information of the keyframe; and outputting the title and the keyframe of the video.
Abstract:
There are provided an apparatus and method for tracking temporal variation of a video content context using dynamically generated metadata, wherein the method includes generating static metadata on the basis of internal data held during an initial publication of video content and tagging the generated static metadata to the video content, collecting external data related to the video content generated after the video content is published, generating dynamic metadata related to the video content on the basis of the collected external data and tagging the generated dynamic metadata to the video content, repeating regeneration and tagging of the dynamic metadata with an elapse of time, tracking a change in content of the dynamic metadata, and generating and providing a trend analysis report corresponding to a result of tracking the change in the content.
Abstract:
Provided is a system for conducting the fast inspection of Android malwares, the system including a processor configured to compute the similarity between the signature for a given target application and one of signatures stored in a database, and a determiner configured to determine whether the target application is a malware based on the computed similarity, wherein the system relates to the technology for examining whether a certain Android application, which can be downloaded via a uniform resource locator (URL), is malicious by examining how similar the application is with the malwares and normal applications verified earlier.