A Real-Time Eye Tracking Based Query Expansion Approach Via Latent Topic Modeling

For most users,it is still considered as a big problem for information retrieval(Retrieval Information)to form a better query to form a better query.Query expansion is an effective method to improve the performance of retrieval.By finding the words and concepts that are related to the original query,and then combined with the original query,the query can provide more positive information to find out the relevant documents from the vast amount of information to improve the user's search experience.Traditional query expansion technology has been in a great extent solve the recall is low,but for precision is difficult to satisfactory results.Personalized query expansion partly solves the problem of low precision.However,the traditional query expansion of the user's search data and it is difficult to capture the user's demand in the query,it is difficult to meet the user's query needs based on user and search engine in real time.Eye Movements can capture the user's gaze information in real time without the user's attention,and then provide users with real-time search behavior data,which is regarded as a new direction of user research and personalized search.Therefore,if the Eye Movements technology is applied in the current personalized query expansion,it will be a new research direction.The main research work of this paper is as follows:First,Eye Movements(Eye Movements)is mainly used in IR are summarized.In addition to the application of Eye Movements in IR,this paper mainly introduces how to use Tracking Eye to capture the user's dynamic search data in real time and how to make use of the eye movement data to make a personalized query expansion.Second,this paper introduces the method of the combination of subject model and Tracking Eye.(Tracking Eye)data can be used to search for the user's potential search intention,which can improve the retrieval results by using the theme model Dirichlet Allocation Latent(LDA).Third,Query Expansion RTQE(Real-Time)is established.Through innovative combination of Eye Movements and LDA,the model can be used to click on a number of documents,record the user's gaze data,the user is refreshed at the current search results or click on the next page of the model to sort and optimize the existing search results by establishing the RTQE model.