Abstract: In accordance with some embodiments of the disclosed subject matter, a method and a system for App page recommendation are provided. The method includes: based on social media contents generated by a user, establishing a parallel corpora comprising an implicit query database including a plurality of user implicit query texts, and an explicit query database including a plurality of user explicit query texts, wherein the plurality of user implicit query texts correspond to the plurality of user explicit query texts; receiving a voice user query from the user; converting the voice user query to a text query; extracting implicit intent text from the text query; inferring a user intent using the parallel corpora based on the extracted implicit intent text; ranking a plurality of App pages according to the inferred user intent; and recommending a subset of the plurality of App pages to the user based on the ranking.

Abstract: In accordance with some embodiments of the disclosed subject matter, a method and a system for App page recommendation are provided. The method includes: based on social media contents generated by a user, establishing a parallel corpora comprising an implicit query database including a plurality of user implicit query texts, and an explicit query database including a plurality of user explicit query texts, wherein the plurality of user implicit query texts correspond to the plurality of user explicit query texts; receiving a voice user query from the user; converting the voice user query to a text query; extracting implicit intent text from the text query; inferring a user intent using the parallel corpora based on the extracted implicit intent text; ranking a plurality of App pages according to the inferred user intent; and recommending a subset of the plurality of App pages to the user based on the ranking.

Abstract: The present invention provides a method for providing augmented product specifications based on user reviews. The method obtains input data of specifications and user reviews on a plurality of products, each specification including at least a pair of a feature and a feature-value of the product. The method concatenates the user reviews of the products to form product-documents, each product-document corresponding to the concatenated user reviews of a product. The method further employs a topic model to process the input data and learn topic distributions in the product-documents and word distributions in topics. The topics include specifications of the products. The topic model is a type of statistical model for discovering topics that occur in a collection of product-documents. Based on the topic model, the method can provide augmented specifications including one or more of relevant sentences of the feature-value, feature importance information, and product-specific words of the product.

Abstract: Provided is a glow plug assembly having dual terminals formed to enable a fast response and reliable control thereof by preventing noise generation caused by grounding, and the assembly including: a housing; a heating tube; a cap; first and second terminals of which lower parts enter inside the heating tube; and a heating coil of which opposite ends are engaged with the first and the second terminals, and a winding part is placed close to the heating tube, wherein the first terminal is shorter than the second terminal in length and is engaged with the heating coil, and the second terminal extends downward to be longer than the first terminal in length, and has a bent part that leads the second terminal to a center of the wound heating coil to secure an insulation space relative to the heating coil, and is engaged with the heating coil.

Abstract: Disclosed is a device capable of preventing a short circuit even in the event of flooding. A terminal polarity fixing unit is disposed between an input terminal unit and an output terminal unit such that a first output terminal and a second output terminal are electrically connected to a neutral terminal and to a phase voltage terminal, respectively, all the time, regardless of how first and second input terminals are paired with the phase voltage terminal and the neutral terminal of an alternating-current outlet. First and second connecting terminals are electrically insulated from each other and spaced apart from each other while being exposed to one side of a body unit of a connecting terminal block made from an insulating body, and electrically connect the first and second output terminals to a load.

Abstract: A method for an app search engine leveraging user reviews is provided. The method includes receiving an app search query from a user, determining a plurality of relevant apps based on the received app search query, and extracting app descriptions and user reviews associated with the plurality of relevant apps from an app database. The method also includes preprocessing the extracted app descriptions and user reviews of each of the plurality of relevant apps to generate a text corpus and creating a topic-based language model for each of the plurality of relevant apps based on the generated text corpus. Further, the method includes ranking a list of relevant apps using the topic-based language model and providing the ranked app list for the user.

Abstract: The present invention provides a method for providing augmented product specifications based on user reviews. The method obtains input data of specifications and user reviews on a plurality of products, each specification including at least a pair of a feature and a feature-value of the product. The method concatenates the user reviews of the products to form product-documents, each product-document corresponding to the concatenated user reviews of a product. The method further employs a topic model to process the input data and learn topic distributions in the product-documents and word distributions in topics. The topics include specifications of the products. The topic model is a type of statistical model for discovering topics that occur in a collection of product-documents. Based on the topic model, the method can provide augmented specifications including one or more of relevant sentences of the feature-value, feature importance information, and product-specific words of the product.

Abstract: A method is provided for ranking media contents. The method includes receiving media contents through a network and extracting feature values of the received media contents. The method also includes implementing a parameter reinforcement learning process to obtain automatically distribution over relativeness and irrelativeness of the received media contents. Further, the method includes ranking the received media contents by a multi-armed bandit algorithm based on the obtained distribution over relativeness and irrelativeness of the received media contents.

Abstract: Embodiments generally relate to systems and methods for generating a sentiment dictionary and calculating sentiment scores of adjectives within the sentiment dictionary. A set of seed words can be identified and expanded using synonyms and antonyms of the set of seed words. Social media data can be parse to identify adjectives that link to the set of seed words with the words “and” or “but.” Matrices representing the attraction and repulsion among the linked adjectives can be generated. A factorization algorithm can be minimized to determine an output matrix that comprises positive and negative sentiment scores for each of the adjectives. In embodiments, a sentiment score for part of all of the social media data can be calculated using the output matrix, and one or more parts of the social media data can be classified as a positive or negative sentiment.

Abstract: Embodiments generally relate to systems and methods for generating a sentiment dictionary and calculating sentiment scores of adjectives within the sentiment dictionary. A set of seed words can be identified and expanded using synonyms and antonyms of the set of seed words. Social media data can be parse to identify adjectives that link to the set of seed words with the words “and” or “but.” Matrices representing the attraction and repulsion among the linked adjectives can be generated. A factorization algorithm can be minimized to determine an output matrix that comprises positive and negative sentiment scores for each of the adjectives. In embodiments, a sentiment score for part of all of the social media data can be calculated using the output matrix, and one or more parts of the social media data can be classified as a positive or negative sentiment.