Research On Intelligent User Interface Based On Bilingual Translation Search Engine

With international communication becoming more and more frequent, the position of translation has promoted unceasingly, the internet search engine-assisted translation obtains the unceasing development. Traditional search engine is based on the words matching approach to retrieve information, but there is polysemant and thesaurus in the natural language of each country, so traditional search engine can't differentiate the semantic information between information, at the same time, the words that user input can't express accurately the user's search intention, resulting in precision of recall of search engine is not high, and displays search engine's mechanicalness, so it can't provides users with good translation effects. The establishments of an intelligent search engine user interface and improve the recall rate become urgent problems.In this thesis, the study is based on a bilingual translation search engine, in view of current search engine's user interface, through establish a synonym dictionary to expand query words, provide input prompt for user to achieve good user experience, enhance the intelligent of search engine user interface.The major ones of this thesis are as follows:First, we analysis the principle of search engine, lexical analysis, the disadvantages of user interface and the methods of query expansion. The effect of local analysis is based on the relevance of the documents that obtained by original query, analysis of click on the search results may obtain the relevance of the result and original query words.Then, using the thought of concept retrieval, establish a synonym dictionary to expand query words, discuss latent semantic analysis to bilingual corpus in order to carry out automatic query expansion, map the index and document in multi-dimensional semantic space, not only to maintain the main message of original data, but also catch the implied potential semantic information, resolve the polysemant and thesaurus problems. We also discuss the establishment of a Chinese and English synonym dictionary and the process of expansion the query words. Analysis the user's click behavior, establish a list of related queries according to this feature, it's convenient for user to adapt query or query expansion.Finally, we use Ajax technology, provide intelligent information to facilitate user input according to user's input when user input query word in text entry box, make the user's operation and the server's response asynchronous, and enhance the user experience. When we retrieval, we expand query words make use of Chinese and English synonym dictionary, enable users to obtain more comprehensive information to translation.