Abstract Recommendation system has been widely used in various types of e-commerce sites. One of the most successful examples is the collaborative filtering algorithm. However, the traditional algorithms only aim at accuracy and ignore these factors closely related with customer satisfaction, such as novelty etc. In this paper, we defined novelty of item from the perspective of the users, designed the... Recommendation system has been widely used in various types of e-commerce sites. One of the most successful examples is the collaborative filtering algorithm. However, the traditional algorithms only aim at accuracy and ignore these factors closely related with customer satisfaction, such as novelty etc. In this paper, we defined novelty of item from the perspective of the users, designed the corresponding offline experiment scheme and evaluation metrics. The dissimilarity and the time-popularity were embedded in the traditional collaborative filtering algorithm, the ability of predicting user's future needs and coverage of recommended list were obviously improved, and the ability of recommended long tail items were also enhanced. Read More Read Less

Abstract Tibetan information processing technology has been obtained some achievements. But it falls behind Chinese and English information processing. It still needs to be paid more attention. Text clustering has the potential to accelerate the development of Tibetan information processing. In this paper, we propose an approach of Tibetan text clustering based on machine learning. Firstly, the approach is to execute... Tibetan information processing technology has been obtained some achievements. But it falls behind Chinese and English information processing. It still needs to be paid more attention. Text clustering has the potential to accelerate the development of Tibetan information processing. In this paper, we propose an approach of Tibetan text clustering based on machine learning. Firstly, the approach is to execute Tibetan word segmentation with Tibetan texts. Then feature selection and text representation are conducted. Finally, K-means and DBSCAN are adopted to deal with the text clustering. The experimental results present that DBSCAN has better performance for Tibetan text clustering. Text clustering systems are designed based on proposed approach. The study is meaningful for the Tibetan text classification, information retrieval as well as construction of high-quality Tibetan corpus. Read More Read Less

Abstract Query expansion adds related words to a user query in order to improve retrieval results. It's an important step in information retrieval. Most of current query expansion methods pay attention to specific expansion strategies or algorithms, while neglecting the query itself. In reaction to the phenomenon, a multistrategy query expansion method based on semantics was proposed. This method started by... Query expansion adds related words to a user query in order to improve retrieval results. It's an important step in information retrieval. Most of current query expansion methods pay attention to specific expansion strategies or algorithms, while neglecting the query itself. In reaction to the phenomenon, a multistrategy query expansion method based on semantics was proposed. This method started by analyzing the semantic structure of user query, and adopted corresponding strategy to select expansion terms. The expansion words are derived from three parts: WordNet, massive web page set and search engine performance evaluation data, which were merged semantically in each expansion algorithm later. The experiment showed this method can improve retrieval results to some extent. Read More Read Less

Abstract With the diversification of customer needs and the widely use of 3D CAD systems, in order to make the 3D CAD system truly become the best carrier of professional technologies and realize the accumulation and reuse of enterprise design knowledge and design process, a design knowledge management method based on 3D CAD system is presented. In this method, a Design... With the diversification of customer needs and the widely use of 3D CAD systems, in order to make the 3D CAD system truly become the best carrier of professional technologies and realize the accumulation and reuse of enterprise design knowledge and design process, a design knowledge management method based on 3D CAD system is presented. In this method, a Design Navigation system based on 3D CAD system is researched, which can build Design Navigations for different products with language engineers are familiar with, and make product design visualized and processed. The mechanism of knowledge classification, management and reuse is also established for Design Navigation, design model, standards manual, formula, charts, drawings, and so on. This method was applied in piston design, and was approved to be effective in design knowledge accumulation. With this method, design experiences of senior staff can be accumulated within the enterprise in forms of parameterized features, standardized design process, precise rules and verified methods. Finally, design efficiency and quality of modularized and serialized products were improved, and the accumulation of design knowledge and design process was strengthened. Read More Read Less

Abstract In the field of real-time data warehousing semistream processing has become a potential area of research since last one decade. One important operation in semi-stream processing is to join stream data with a slowly changing diskbased master data. A join operator is usually required to implement this operation. This join operator typically works under limited main memory and this memory... In the field of real-time data warehousing semistream processing has become a potential area of research since last one decade. One important operation in semi-stream processing is to join stream data with a slowly changing diskbased master data. A join operator is usually required to implement this operation. This join operator typically works under limited main memory and this memory is generally not large enough to hold the whole disk-based master data. Recently, a seminal join algorithm called MESHJOIN (Mesh Join) has been proposed in the literature to process semistream data. MESHJOIN is a candidate for a resource-aware system setup. However, MESHJOIN is not very selective. In particular, MESHJOIN does not consider the characteristics of stream data and its performance is suboptimal for skewed stream data. In this paper we propose a novel Semi-Stream Join (SSJ) using a new cache module. The algorithm is more appropriate for skewed distributions, and we present results for Zipfian distributions of the type that appears in many applications. We present the cost model for our SSJ and validate it with experiments. Based on the cost model we also tune the algorithm up to a maximum performance. We conduct a rigorous experimental study to test our algorithm. Our experiments show that SSJ outperforms MESHJOIN significantly. Read More Read Less