Search

The paper presents a novel approach for customer segmentation which is the basic issue for an effective CRM (Customer Relationship Management). Firstly, the chi-square statistical analysis is applied to choose set of attributes and K-means algorithm is employed to quantize the value of each attribute. Then DBSCAN algorithm based on density is introduced to classify the customers into three groups (the first, the second and the third class). Finally biclustering based on FP-tree algorithm is used in the three groups to obtain more detailed information. Experimental results on the dataset of an airline company show that the biclustering could segment the customers more accurately and meticulously. Compared with biclustering based on Apriori, the Fp-tree is more efficient on the large dataset.

Discovering and matching services among heterogeneous agents is the fundamental process for agent cooperation and coordination in multi-agent systems. In such a process, numerous service-providing agents are matched against requested criteria. During the matching procedure, an agent is given the chance to discover its new corresponding agents that can provide desirable services. However, in order to solve more complex problems, the number of agents in a multi-agent system is increasing rapidly; therefore making the chance for an efficient and desirable match is decreased. Thus, improving the efficiency of the agent matching process has become an important issue in the multi-agent filed. Utilising an appropriate agent-matching mechanism will enhance agent cooperation and communication efficiency in dealing with complex problems. In this chapter, we develop a new agent-matching algorithm, the Agent-Rank (AR hereafter) algorithm, which ranks service-providing agents according to their contributions to a nominated requesting agent. The AR algorithm improves agent matching process through combining the general ranking scores with the request-based ranking scores. The AR algorithm narrows down the searching targets (services), which is avail to detect some random targets in a chance discovery process.

Human action recognition from video has attracted great attentions from various communities due to its wide applications. Regarded as an effective way to analyze human movements, human skeleton is extracted and represents human body as dots and lines, Recently, depth-cameras make skeleton tracking become practical. Based on the extraction and template matching, we develop a system for online human action segmentation and recognition in this paper. We proposed a method to generate action templates that can be used to represent intra-class variations. We then adopted efficient subsequence matching algorithm for online process. The experimental results demonstrated the effectiveness and efficiency of our system.

Making decisions in a dynamic environment is complicated and indefinite because of various unpredictable factors and large volumes of information. It has become a key issue for modern organizations to have efficient systems and tools to support rational decision-making in problem-solving processes, particularly, for problems with a lack of predefined structure. This paper provides a novel agent-based framework for Decision Support Systems (DSS) based on the analysis of the major problems in existing DSS. The proposed multi-agent-based framework, namely the Agent-Matrices framework, provides an open, efficient and flexible architecture for DSSs. The Agent-Matrices framework overcomes the compatibility and connectivity problems in most traditional DSS applications. This framework utilises Matrices to allow agents to acquire information, self-upgrade, perform tasks, travel to other Matrices, and be reused. This paper primarily introduces the methods used for coordinating the Matrices and agents to solve a complex problem.

As a new tourism form, the urban tourism is rising gradually in the world. The brand image of urban tourism can divide into four dimensions, which were regional image, landscape image, cultural image, and service image. The survey results from 97 non-local tourists showed that the perceptions of “regional image” and “cultural image” were strongest, but perception of “landscape image” was weakest. Except the landscape image, the perception of urban tourism brand image had remarkable influence on the choice of tourism destination. So the government shall strengthen environmental construction and carry out integrated marketing communication to promote urban tourism brand image in Ningbo.

CAD/CAM software products can help boost productivity for machining medical parts. However, the process of evaluating and re-calculating CAD model design is basically carried out manually. The demand for automated CAD process systems has been rising. Automated feature recognition (AFR) systems can improve system efficiency and effectiveness for processing CAD models in manufacturing sectors, particularly for designing medical machining parts. However, existing AFR methods are unable to fulfill industrial requirements for extracting and recognizing domain components from CAD models efficiently. In this paper we suggest a knowledge-based AFR system that can efficiently identify domain components from CAD models. The AFR knowledgebase incorporates rule-based methods for identifying core components from CAD models. The process of defining the rules and fact base structure is one of the most critical issues in the AFR system design. There is no existing technology available for generating inference rules from the STEP model format. The AFR-based system has successfully solved the technical issues in both the inference process and STEP-based extraction process. The skeleton software has been successfully developed based on the modularized system framework. The skeleton software can effectively recognize the common domain specific components.

This book constitutes the refereed proceedings of 4 international workshops held in conjunction with the 14th Asia-Pacific Web Conference, APWeb 2012, in Kunming, China, in April 2012 (see LNCS 7235). The 29 revised full papers presented were carefully reviewed and selected for presentation at the following 4 workshops: the 1st workshop on sensor networks and data engineering (SenDe 2012), the 1st international workshop on intelligent data processing (IDP 2012), the 1st international workshop on information extraction and knowledge base building (IEKB 2012), and the 3rd international workshop on mobile business collaboration (MBC 2012).

Existing research work on intelligent agents is facing the challenge of extending agent capabilities. One of major difficulties of applying intelligent agents in various software platforms lies in that generalized agent design models may not fit with specified organizational applications. This problem dramatically hampers the practical applications of intelligent agents. In this paper we point out several aspects that might affect the development of agent-based technology; and further address the information reuse and unification issues through suggesting an enhanced agent capability reuse mechanism, which is able to provide an efficient process for agent capability reuse. The proposed design can fulfill the needs for a dynamic agent environment.

On the foundation of analyzing the existing classification, an acquisition method of extension transformation knowledge based on Decision Tree classification has been proposed The new-bored method re-mines and transforms the decision tree rules to "can't to can, not to yes" strategy which aims to provide targeted decision-making on the transformation of the customer churn by flexible use of the extension set and extension transformation theory. Its practice in a web company has proved that this method is highly feasible, and also has the reference value for other methods research based on Extenics.

Physiological and genetic information has been critical to the successful diagnosis and prognosis of complex diseases. In this paper, we introduce a support-confidence-correlation framework to accurately discover truly meaningful and interesting association rules between complex physiological and genetic data for disease factor analysis, such as type II diabetes (T2DM). We propose a novel Multivariate and Multidimensional Association Rule mining system based on Change Detection (MMARCD). Given a complex data set ui (e.g. u1 numerical data streams, u2 images, u3 videos, u4 DNA/RNA sequences) observed at each time tick t, MMARCD incrementally finds correlations and hidden variables that summarise the key relationships across the entire system. Based upon MMARCD, we are able to construct a correlation network for human diseases.

Existing multi-agent systems lack a comprehensive methodology for agents to perceive the structures of a network, which they are residing in. This problem obstructs efficient agent communication and transmission within a multi-agent system. The demand for an efficient mechanism allowing an agent to sense other agents' locations has been growing substantially, since agents are mobile and cooperative entities. In this paper, we propose a new Topological Description Language for Agent networks, namely TDLA, to provide agents with the capability of understanding their environment. TDLA helps an agent perceive the network structure that it resides in. It adopts traditional topological analysis methods for local area networks and further tailored for agent networks.

Modern engineering design, analysis and manufacturing activities rely heavily on software to handle increasing volumes of data and model complexity. Automated Feature Recognition (AFR) technologies are highly demanded by manufacturing sectors since AFR can efficiently improve the performance of Computer-Aided Design (CAD) processes and reduce costs. Nevertheless, most existing FR applications are confronting various problems of processing CAD models in the manufacturing industry, such as aerospace and automobile industries. The missing link between CAD models and knowledge-based tools is one of the major obstacles. This research project investigates the feasibility and benefits of bridging the gap between knowledge based mechanisms and CAD models, and suggests a knowledge-based AFR approach for tackling AFR problems occurring in the computer-aid manufacturing design process. The AFR system significantly reduces time and costs of analysing CAD models for downstream design processes.