Abstract Most of data about an item in recommendation algorithms actually consists of meta-information. We argue that meta-information heavily affects the recommendation quality. This paper analyses the impact of deeper representations and more elaborate similarity measures in the quality of the recommendation of music. Firstly, we propose a innovative approach to extract musical features from audio files and use it to deepen song representation.... Most of data about an item in recommendation algorithms actually consists of meta-information. We argue that meta-information heavily affects the recommendation quality. This paper analyses the impact of deeper representations and more elaborate similarity measures in the quality of the recommendation of music. Firstly, we propose a innovative approach to extract musical features from audio files and use it to deepen song representation. Next, we propose a newer metric for measuring diversification. Finally, we evaluate both diversification and accuracy for the recommendation of songs. Results show that the inclusion of these deeper features on the representation of the item improves the accuracy and greatly improves diversification. Read More Read Less

Abstract Discretization of real value attributes is a vital task in data mining, particularly in the classification problem. Discretization part is also the crucial part resulting the good classification. Empirical results have shown that the quality of classification methods depends on the discretization algorithm in preprocessing step. Universally, discretization is a process of searching for partition of attribute domains into intervals and unifying the... Discretization of real value attributes is a vital task in data mining, particularly in the classification problem. Discretization part is also the crucial part resulting the good classification. Empirical results have shown that the quality of classification methods depends on the discretization algorithm in preprocessing step. Universally, discretization is a process of searching for partition of attribute domains into intervals and unifying the values over each interval. Significant discretization technique suit to the Intrusion Detection System (IDS) data need to determine in IDS framework, since IDS data consist of huge records that need to be examined in system. There are many Rough Set discretization technique that can be used, among of them are Semi Naives and Equal Frequency Binning. Read More Read Less

Abstract In order to solve the problems of poor intelligence, interactivity and adaptability of Japanese teaching system, this paper designs a kind of high autonomy Japanese digital learning system based on Agent technology. The system includes 8 intelligent Agent, 4 databases and 2 knowledge inventory. It completes the interaction between learners, Japanese teachers and data systems through the collaboration between multiple Agents, and investigates... In order to solve the problems of poor intelligence, interactivity and adaptability of Japanese teaching system, this paper designs a kind of high autonomy Japanese digital learning system based on Agent technology. The system includes 8 intelligent Agent, 4 databases and 2 knowledge inventory. It completes the interaction between learners, Japanese teachers and data systems through the collaboration between multiple Agents, and investigates the effectiveness of the system. The results show that the application of independent Japanese digital learning system has high efficiency, flexibility and autonomy, which is of great significance for the future study of Japanese. Read More Read Less