Abstract: A machine reading comprehension apparatus and method are provided. The apparatus receives a question and a text. The apparatus generates a plurality of first predicted answers and a plurality of first source sentences corresponding to each of the first predicted answers according to the question, the text, and the machine reading comprehension model. The apparatus determines a question category of the question. The apparatus extracts a plurality of special terms related to the question category and a plurality of second source sentences corresponding to each of the special terms from the text. The apparatus concatenates the question, the first source sentences, the second source sentences, the first predicted answers, and the special terms into an extended string. The apparatus generates a plurality of second predicted answers corresponding to the question according to the extended string and the micro finder model.

Abstract: A machine reading comprehension method includes the following operations: performing a relation augment self attention (RASA) feature extraction process on at least one historical dialogue data and a current question data respectively to obtain at least one historical dialogue feature and a current question feature; and performing a machine reading comprehension (MRC) analysis according to the at least one historical dialogue feature and the current question feature to obtain a response output.

Abstract: A machine reading comprehension method is disclosed. The machine reading comprehension method includes the following operations: performing a relation augment self attention (RASA) feature extraction process on at least one historical dialogue data and a current question data respectively to obtain at least one historical dialogue feature and a current question feature; and performing a machine reading comprehension (MRC) analysis according to the at least one historical dialogue feature and the current question feature to obtain a response output.

Abstract: An apparatus and method for generating a dialogue state tracking model. The apparatus retrieves a field feature corresponding to a queried field from a database according to the queried field corresponding to a queried message. The apparatus retrieves a candidate-term feature corresponding to each of at least one candidate-term corresponding to the queried field from the database, and integrates them into an integrated feature. The apparatus also generates at least one relation sub-sentence of a reply message corresponding to the queried message and generates a sentence relation feature according to the at least one relation sub-sentence. The apparatus further generates a queried field related feature according to the field feature, the integrated feature and the sentence relation feature and trains the dialogue state tracking model according to the queried field related feature.

Abstract: An apparatus and method for generating a dialogue state tracking model. The apparatus retrieves a field feature corresponding to a queried field from a database according to the queried field corresponding to a queried message. The apparatus retrieves a candidate-term feature corresponding to each of at least one candidate-term corresponding to the queried field from the database, and integrates them into an integrated feature. The apparatus also generates at least one relation sub-sentence of a reply message corresponding to the queried message and generates a sentence relation feature according to the at least one relation sub-sentence. The apparatus further generates a queried field related feature according to the field feature, the integrated feature and the sentence relation feature and trains the dialogue state tracking model according to the queried field related feature.

Abstract: A semantic analysis method, semantic analysis and non-transitory computer-readable medium are provided in this disclosure.

Abstract: A semantic analysis method, semantic analysis and non-transitory computer-readable medium are provided in this disclosure.

Abstract: A method for generating a conversational user interface is provided. The method is executed by a processor to perform the following steps. Extract the content of a webpage. Obtain a first form in the webpage. Obtain multiple first fields according to the first form. Create a first slot filling model based on the multiple first fields. Generate a conversational user interface. The processor is configured to use the first slot filling model to determine data required by the first fields, and obtain data that can be filled in the first fields according to an input dialogue data to fill in the first fields.

Abstract: A drug combination prediction method comprising: storing a plurality of original gene sets, at least one first gene impacted by a first drug and at least one second gene impacted by a second drug; determining the part of the at least one first gene and the part of the at least one second gene to be a first interaction gene set; calculating a gene amount of the first interaction gene set to obtain a first interaction gene amount, and calculating a first percentage generated by the first interaction gene amount in the first original gene set; calculating an interaction value of the combination of the first drug and the second drug according to the first percentage; and selecting at least one synergistic pharmaceutical composition according to the interaction value.

Abstract: A conversation analysis method includes: receiving, by a processor, a conversation data including a plurality of sentences sorted by time; performing, by the processor, distributional clustering of context vectors to a plurality of words shown in the sentences to obtain a word order between the words; analyzing, by the processor, the words shown in the sentence to obtain a basic conversation matrix according to the word order; performing, by the processor, a fuzzy matching to the basic conversation matrix to obtain a conversation matrix based on the basic conversation matrix; detecting, by the processor, a topic trend according to the conversation matrix to determine the topic of the conversation data; and outputting, by the processor, the conversation matrix and the topic trend corresponding to the conversation data to a database.

Abstract: A de-identification data generation apparatus, method, and non-transitory computer readable storage medium thereof are provided. The apparatus is stored with a plurality of original records, wherein each of the records has a plurality of original values corresponding to a plurality of attributes one-to-one. The apparatus decides a plurality of attribute relations (including a user-defined attribute relation) according to the original values, wherein each attribute relation is defined by two attributes. The apparatus decides a plurality of relation groups according to the attribute relations. For each relation group, the apparatus calculates a statistical distribution of the original values corresponding to the attributes in the relation group, aggregates the statistical distribution into a plurality of sub-statistical distributions, and adds noise to each sub-statistical distribution individually.

Abstract: A webpage data extraction device and a webpage data extraction method thereof are provided. The webpage data extraction device operates to: group the webpage data into URL groups according to URL relation of the webpage data; select a first webpage data and a second webpage data from each URL group; analyze the first webpage data and the second webpage data to derive a webpage node data set; group webpage node data of the webpage node data set into webpage node data groups according to XPath relation and text content relation of webpage node data of the webpage node data set; calculate respective text content sum for each webpage node data group; determine main webpage node data groups from the webpage node data groups according to the text content sums; decide main content extraction information based on XPath of the webpage node data included in the main webpage node data groups.

Abstract: The present disclosure illustrates a website simplifying method including following steps: acquiring a website having a plurality of web pages; analyzing Uniform Resource Locator (URL) addresses of the web pages; establishing a website-structure in accordance with the URL addresses of the web pages; and reestablishing guiding links of the web pages to generate a criss-cross browse-associating table. The website-structure has a plurality of website-structure levels in which respectively have at least one of web pages. The criss-cross browse-associating table records reestablished guiding links of the web pages and a criss-cross structure is formed by the reestablished guiding links of the web pages.

Abstract: A next generation sequencing analysis system and a next generation sequencing analysis method thereof are provided. The next generation sequencing analysis system receives a target gene input, and decides at least one gene group of the target gene input based on gene related information stored in a gene database.

Abstract: A method for recommending tour attractions based on medical services includes steps of choosing at least one medical service from a plurality of medical services; calculating suitability between the at least one medical service and tourist attractions based on medical degree parameters and tour degree parameters; and recommending at least one tourist attraction from a plurality of tourist attractions based on the suitabilities. Furthermore, a system for recommending tourist attractions based on medical services is also disclosed herein.