With the rapid development of information technology, the concept of big data is used in information collection on different things, especially for the text classification. This paper propose an improved KNN algorithm based on clustering for the automatic classification of Web text. In addition, we find a new method to find out which text in the same category belongs to the same cluster. Finally, we classify Web text automatically and test them by using the existing and improved KNN algorithm respectively. Simulation results show that the improved algorithm can significantly raise the accuracy of automatic classification.

Application of kNN improved algorithm in automatic classification of network public proposal cases

User interest modeling is the key technology of personalized information retrieval. The access behavior, which user browse, save, collect, or print Web page, can really represent user interest. This paper expatiates on the relation between user interest and access behavior on the basis of vector space model and document vectorization, discusses the quantificational formula of user interest degree which is affected by document structure or time forgetting factor, and introduces user interest modeling algorithm. The system structure, result filtration, user interest model maintenance of the personalized information retrieval system based on multi-agent are presented in detail. The experimental result indicates that the system can truly reflect user interest, effectively improve the precision rate and recall ratio of information retrieval, and obviously increase user’s satisfaction degree.

User Interest Modeling Based on Access Behavior and its Application in Personalized Information Retrieval

This paper introduces the definition and characteristics of vertical search engine briefly, designs an application model of vertical search engine to improve the efficiency for users' searches and describes the algorithm of the key module of the model in detail. The experiment, using the advanced algorithm to compare the traditional search engine and vertical search engine, proves the feasibility of algorithm and practicability of application model.

Design of an Application Model Based on Vertical Search Engine

At present, people no longer meet the way of communication between users and the Internet. And more and more people choose the interaction between users and users to get information. The community question answering system is one of the new information sharing model. In the community question answering system, users are not only the questioner but also the answer and the question is the link between the users. With the increasing number of users and the increasing number of questions and answers, it makes many questions which just were raised disappear in the category pages of the home page. Leading to the efficiency of the questions be answered greatly reduce. Aim at the recommended user’s interest, ability and time. In this paper we construct a dynamic user interest model and user expertise model. Experimental results show that the recommendation mechanism improves the efficiency of the recommendation to a certain extent.

Question Recommendation Based on User Model in CQA

This paper builds important evaluation indexes for urban rail transit network. Besides degrees and betweenness in complex network, I add PageRank value and passenger flow indicators in consideration of the urban rail transit network topology model to better evaluate relationships between stations in urban rail transit network and offer theoretical guarantee for risks. We provide factor analysis method to calculate the importance degree of each station in the networks. Then illustrated by the case of Beijing Railway, and we compare the results with K-means cluster analysis. The results show that the factor analysis and the cluster analysis has consistence characteristics in macroscopic, but factor analysis can reflect the over-rall properties of indicators better and possess the explicit presentation. It has better actual leading meanings.

Factor analysis of key nodes in urban rail network

Based on the study of the search engine key technology, we analyze the PageRank algorithm and propose an improvement aimed at the deficiency of the PageRank algorithm

Improvement of PageRank Algorithm for Search Engine

With the development of the Co-operative education in China, the Co-operative Studies Supporting System (CSSS) for Software Engineering becomes a task of high priority. We analyze the requirements of integrated study and work term of Co-operative education in software engineering, give a new Integrate Developing Model based on Capability Maturity Model (IPD-CMM) supporting by Team Software Process (TSP) and Personal Software Process (PSP) methodology. Then, the system design is given and functions are introduced with Data Collection Planning, Tracking, Data Analysis and Data Export. In the end, a case study is shown with customized workflow and the postmortem analysis is shown including Defect analysis, Plan analysis, Process analysis and Quality analysis. CSSS can carry on a quality evaluation on every part of the software and even make an evaluation and an analysis on the entire course of Cooperative education. It increases the efficiency, reduces or eliminates the unnecessary process that is convenient to manage and lessen the interior consumption to a minimum.

Zhang Yan-hong

Papers

Improvement of PageRank Algorithm for Search Engine

Co-operative Studies Supporting System for software engineering