Search

The need for computer-based intelligent techniques for recruitment and retention of employees in a highly competitive global market has grown significantly in the last decade. Salesperson recruitment is a critical task for most organisations. Existing approaches for salesperson recruitment primarily rely on filtering of applications based on selection criteria followed by interviews. Some organisations also include personality testing based on psychometric techniques. The high turnover of salesperson in the industry suggests limited success of these procedures. Additionally, existing approaches lack benchmarking methods. In this paper we describe design and development of an intelligent sales recruitment and benchmarking system (ISRBS) for recruitment and benchmarking of salespersons. ISRBS design represents operation of the findings and outcomes based on actual field studies and random surveys of salespersons as well as development of models for measuring independent and dependent variables related to selling behaviour. The main contributions of the paper are (i) Developing an on line selling behaviour profiling technique based on integration of intelligent system techniques like expert systems and fuzzy sets, psychology based selling behaviour model, and AHP techniques, and (ii) an objective and novel selling behaviour benchmarking technique to facilitate modelling of organisation based benchmarks and cultural fits. An earlier version of this system has been commercially used in the industry in Australia. ISRBS integrates psychology based selling behaviour model with artificial intelligence techniques and soft computing methods for selling behaviour profiling and benchmarking.

Currently, TV programs are evaluated by using questionnaires given after previews or by using TV ratings. There are few objective criteria useful for describing technical know-how about program production. One of the TV program producers' concerns is how to choose expression methods that convey their ideas to viewers correctly and efficiently. Research has shown that eye-gaze direction is related to the human focus and attention. Gaze-based evaluations have been proposed for image-quality evaluations and certain usability tests. Such approaches are mainly based on how often a specific region attracted the subjects' gaze or how long their gaze was fixed on it. To apply these approaches to TV programs, all the object regions that seem to attract a viewer's gaze need to be specified in advance. This causes several problems including the accuracy of specifying the region by using an image processing technique is not equal to the human subject's recognition ability and it is not feasible to manually specify such regions in an enormous number of frames (images) comprising the program. Further, how characteristics of well-produced TV programs appear on the viewer's gaze has not been objectively analyzed yet. There is a need to investigate the relationship between gaze and program contents which can be used as means for improving comprehension and quality of the TV programs. In this paper, we propose a new measurement and evaluating method for this purpose. This paper focuses on the relationship between a viewer's comprehension of a program and their gaze direction in a real experimental TV educational program involving 26 elementary school children and broadcast by NHK Broadcasting Corporation of Japan. Correlation between TV program comprehension and entropy is investigated. That is, variances in the gaze direction in relation to program comprehension are based on a entropy value that represents the degree of dispersion in each frame and is calculated from a probability density function estimated from the gaze directions. The results indicate that the variances of the gaze direction for scenes that gave better comprehension tended to be lower. This tendency was further noticeable after a keyword utterance were related to the answers of corresponding questions.

This paper demonstrates an innovative application in the context of autism using a Humanoid robot. A nine-month long home-based care trial with 191 hours of interaction is conducted to investigate the engagement with two young people with autism through adaptation on service design (e.g., change from narrative voice to more interactive story telling using emotion expressions, sounds effects and background music) and service personalization (service novelty). It involves big data collected to learn the participants' preferences and activity patterns. The experimental evidence on how to improve the long-term benefits of use of social robots is reported. The research establishes that in home-based care a Humanoid robot can maintain their engagement with young people with autism through service personalization over long periods of time.

The 90’s has seen the emergence of hybrid configurations of four most commonly used intelligent methodologies, namely, symbolic knowledge based systems (e.g. expert systems), artificial neural networks, fuzzy systems and genetic algorithms. These hybrid configurations are used for different problem solving tasks/situations. In this paper we describe unified problem modeling language at two different levels, the task structure level for knowledge engineering of complex data intensive domains, and the computational level of the task level hybrid architecture. Among other aspects, the unified problem modeling language considers various intelligent methodologies and their hybrid configurations as technological primitives used to accomplish various tasks defined at the task structure level. The unified problem modeling language is defined in the form of five problem solving adapters. The problem solving adapters outline the goals, tasks, percepts/inputs, and hard and soft computing methods for modeling complex problems. The task structure level has been applied in modeling several applications in e-commerce, image processing, diagnosis, and other complex, time critical, and data intensive domains. We also define a layered intelligent multi-agent, operating system processes, intelligent technologies with the task structure level associative hybrid architecture. The layered architecture also facilitates component based software modeling process.

Like most of the developed countries, Australia's population is ageing. In this paper the authors report on deployment of a socially assistive robot (Matilda) based multi-layer service architecture to support People with Dementia (PwD) in home-based care. The robots deliver a range of services in an emotionally engaging manner based on the lifestyle of PwD. The assistive social robots have been deployed in several Australian households over a six-month period. The results of trial demonstrate that multi-layer service architecture with personalized services and human-like communication modalities of Matilda have the ability of breaking technology barriers, providing sensory enrichment and social connectivity to the PwD as well as augmenting their good memories and providing respite to the partners.

In this paper, the learning knowledge and strategy of a generic neuro-expert system (GENUES) architecture for the training of neural networks in a power network alarm processing system is described. The GENUES architecture forms an important part of an integrated architecture used for developing a real-time alarm processing system in a regional power system control centre. The authors also report on some of the important implementation issues related to alarm processing.

This research aims to contribute to the body of knowledge regarding the use of assistive robots for tele-health care in the home through an innovative service model. In contrast to existing tele-consultation, tele-monitoring and tele-medicine service models, the assistive robot enabled tele-health model focuses on providing holistic care (which includes reactive, proactive as well as preventative care). Additionally, unlike existing tele-health technologies it facilitates human-like engagement and reciprocal association between assistive robot and the human partner.

This paper describes the design of multimodal robotic system, embodiment of multimodal interaction (voice, gestures, emotion, touch panel and dance) in assistive social robot (Matilda) for modeling group based and one-to-one interactions with technology adverse elderly in nursing homes. It describes the human-centered evaluation of Matilda based on quality, naturalness, user satisfaction and predictive accuracy based on first ever field trials in Australia. The multimodal social robots have facilitated breaking the technology barriers with the elderly leading to several aged care facilities and community centers showing interest in future trials.

This paper describes the role and correlation of non-verbal data (emotions) with other data in managerial decision making in natural environment. The work is targeted towards context-aware emotionally intelligent Human Resource Management (HRM), information personalization and managerial decision making in consumer and service sector industries like finance/banking, electronics, retail, tourism and health care.

In this paper we outline a seven layer context-aware data mining architecture which combines context, sensemaking (cognitive and affective) and data mining technologies to design adaptive context-aware data mining systems. We particularly show how cognitive constructs and emotional attitudes of a user mediate in interpretation of meaning in hidden patterns. We illustrate the role of cognitive constructs in interpreting a CRM situation by a relationship manager in a banking and finance application. We also illustrate the role of emotional attitudes as an important factor in context-aware interpretation of mined behavioral patterns in a sales recruitment and benchmarking application.

In this paper we describe a novel intelligent web based system which employs human feedback and evaluation to assist an Internet user to interactively identify the clothing of a missing person. It is novel in terms of employing GA to interactively create different shirt designs. It also illustrates human-centeredness from two perspectives. Firstly, from a multimedia information system perspective, in order to capture the user semantics it employs user feedback. Secondly, from a computational intelligence perspective the user is involved in defining objective function of a GA agent for optimizing search. The system can be used by the law enforcement authorities to identify the type, color and design of the shirt worn by a missing person. It has been tested on a small number of users with encouraging results.

Emotions form an important component of human behavior and decision making. In this paper we employ image processing and soft computing techniques to design and implement emotional state profiling of a sales candidate while they provide psychological inputs related to their selling behavior to an intelligent sales recruitment and benchmarking system. This research is part of a larger effort aimed at giving computers and other pervasive devices the skills of emotional intelligence, such as the ability to recognize a person's emotions and to respond appropriately to those emotions. The work reported has implications in not only in recruitment and benchmarking but also web recommender and information personalization systems.

This paper proposes a human-centered multi-layered multi-agent architecture for performance optimization of data mining applications. Feedback from user is used for optimizing the data mining performance. This paper describes application of the architecture in banking and finance domain

E-recruitment area has emerged as an important e-business function in the past few years. We describe a novel e-recruitment multiagent application for recruitment and benchmarking of salespersons. Most existing approaches to recruitment rely on the interview process or on psychometric techniques. Both these approaches have had limited success. We describe a multiagent e-sales recruitment system (e-SRS), which integrates a selling behavioral model with expert systems and soft computing techniques like fuzzy-K-means for predicting the selling behavior profile of a sales candidate.

Recently in the world, many researches of personal identification method using biometrics are widely done. Especially, personal identification using the face is used because of needless for physical contact. In this paper, we propose the new method to identify the personal using the SPCA and the NN. The SPCA has been proposed in order for the PCS to achieve high-speed processing. Moreover, the personal identification system needs to have not only ability to recognize registrant correctly but also ability to reject un-registrant certainly. Therefore, we perform simulation using un-registrant. Furthermore, we analyzed and examined about the individual feature in a face by using the GA and the SPCA. Then, in order to show the effectiveness of the proposed method, we show computer simulations by suing the real image. From these results, we show the effectiveness.

Identifying profiles of customers based on their history trails of transactions and further providing personalization services to them are the best strategy to retain customers for online businesses. Currently profiling techniques have the defects of static offline operation, inexactness, and slow response, which do not meet the ever-changing demand of online transactions for e-commerce, e-banking, etc. This paper proposes a dynamic intelligent profiling architecture with the help of high performance computing. The architecture is tested and applied in an e-banking application.

Intelligent agents (or software agents) today can be seen as providing support to practitioners and problem solvers at three levels, namely, clerical, tool and task levels. In particular, we outline computational level based agent definitions of fuzzy neural agents. We show the application of these agents in determining faulty components in a electrical power system network.

A generic neuro-expert system architecture which can overcome difficulties faced by stand-alone expert systems and artificial neural networks is proposed. It can be applied in various problem domains, such as engineering and fault diagnosis, which require problem decomposition. It is recommended for use in real-time systems. The neuro-expert system architecture can be used at different levels of a power system hierarchy for alarm interpretation and fault diagnosis. >

Electronic Commerce (EC) can be broadly seen as the application of information technology and telecommunications to create virtual trading networks where goods and services are sold and purchased.

The applications in this book employ a number of concepts and technologies associated with e-business. The e-business concepts include types of e-business systems, e-business strategies and e-business models. The technologies can be grouped under Internet and web technologies, intelligent technologies, web software engineering (Object-oriented and agents) technologies, and multimedia. E-business strategies and e-business models also provide the context in which e-business technologies are applied by organizations to gain competitive advantage. This chapter introduces the reader to these concepts and technologies.

In the last chapter we outlined relevance feedback as one of the methods for developing user-centered multimedia applications. On the other hand, researchers in the computational intelligence or soft computing community have been recently trying to develop intelligent applications which humanize computational intelligence technologies (Takagi 2001, 2002). In this chapter we describe an intelligent web multimedia system which employs relevance feedback as a means of assisting an Internet user (relative or friend of a missing person) to interactively identify the clothing of a missing person. The system can be used by the law enforcement authorities, like the police to identify the type, color and design of the shirt worn by a missing person.

The objective of this chapter is to outline the computational framework of intelligent multimedia multi-agent systems based on the human-centered approach. The title human-centered virtual machine encapsulates the integration of conceptual components of the human-centered system development framework and the technology based artifacts used to realize the conceptual components at the computational level.

In this paper we outline a multi-layered multi-agent distributed soft computing architecture for providing support to practitioners and problem solvers at four levels, namely, distributed processing, technology, optimization and problem solving level respectively. The architecture has been applied in complex problems like image recognition, web mining, alarm processing, and medical diagnosis. We illustrate some aspects of the architecture with help of e-Sales recruitment (e-SRS) application. An earlier version of e-SRS (expert system ) is also in commercial use.

Internet Personalized services are irresistible developing trend for e-commerce. More and more researchers are committed to personalization field. Many personalization approaches are static and lack of means to improve the personalized tasks. This paper proposes an adaptive e-commerce personalization framework using traditional data mining techniques and agent technology as well as user feedback optimisation mechanism to improve the personalized services to the e-commerce customer. The behaviours of all the agents in the framework are carefully considered and the framework has been applied to an online banking system.

In this paper we outline a multi-layered distributed agent ontology for providing system modeling support to practitioners and problem solvers at four levels, namely, distributed, tool (technology), optimization and task level respectively. We describe the emergent characteristics of the architecture from a architectural viewpoint. We also outline how the ontology facilitates development of human-centered soft computing systems. The ontology has been applied in a range areas including alarm processing, web mining, image processing, sales recruitment and medical diagnosis. We also outline definition of agents in different layers of the ontology.

An integrated model for real time alarm processing in a real world terminal power station is applied. The integrated model is a combination of a generic neuro-expert system model, object model, and UNIX operating system process (UOSP) model. It is shown how the massive parallelism and fast execution features of ANNs help to cope with real-time system constraints like data variability and fast response time. For further enhancing reliability, a practical use of competing expert system-artificial neural networks (ES-ANN) objects is proposed. >

We outline the features of a cognitively compatible symbolic-connectionist architecture for complex data intensive time-critical domains especially in the diagnostic area. We examine how a symbolic-connectionist architecture copes with the time-critical and size complexity aspects of a real time alarm processing system developed for a regional power system control centre. >

In this paper we describe the problem-solving constructs, namely information processing constructs, learning constructs, knowledge representation constructs and computational constructs for building complex symbolic-connectionist systems. These constructs are applicable in particular to complex diagnostic domains and in general to complex data intensive domains. >

This paper outlines a human-centered virtual machine of problem solving agents, intelligent agents, software agents and objects. It deals with issues related to high-assurance (e.g. reliability, availability, real-time and others) through design of human-centered system architecture in which technology is a primitive. The human-centered virtual machine is based on a number of human-centered perspectives including the distributed cognition approach. The human-centered virtual machine has been applied in complex data intensive time critical problems like real-time alarm processing and fault diagnosis, air combat simulation and business (decision support).

Intelligent Multi-Agent Hybrid Distributed Architecture (IMAHDA) for complex systems has been defined at the task structure level and the computational level. IHMADA has been built around five information processing phases. The five phases represent a deliberate as well as automated or reflexive reasoning structure. This paper outlines some aspects of the computational level of IMAHDA. The computational level is derived from four layers, namely, object layer, software agent layer, intelligent agent layer, and problem solving agent layer. Agent description in terms of percepts, goals, actions, and environment for some of the problem solving agents is also outlined. IMAHDA has been applied in engineering (real time systems) and management areas.

Knowledge based systems, fuzzy systems, and artificial neural networks (ANN) are the three most widely used computational paradigms for emulating different aspects of human cognition like information processing, knowledge representation, and learning. A proper integration of these three paradigms can help to realize powerful problem solving strategies especially for large data intensive domains. In this direction we have developed a generic architecture for integration of symbolic (knowledge based and fuzzy) and connectionist (ANN) systems for large, data intensive domains at the task structure level, computational (symbol) level, and the program level. In this paper we outline the knowledge modeling aspects of the integrated symbolic (knowledge based and fuzzy)-connectionist architecture. The knowledge content of the architecture can facilitate a problem solver in modeling the knowledge required for using as well as integrating the three intelligent paradigms.

Intelligent agents are being increasingly used on the Internet to provide various kinds of support to the Internet users. Task level support is aimed at modeling the user's tasks and problem solving models. In this work, we focus on design and application of intelligent agents for providing task level support in Internet-based applications via an Electronic Broker. Our Broker is used to locate the desired information or product for its user and acts as a mediator between the information providers or on-line suppliers and user/customer. The operation of the intelligent electronic broker is demonstrated through a sample application, namely, the purchase of hardware adapters on the Internet.

In the last two chapters we have looked into the application of four components of the HCVM. In this chapter we wish to demonstrate two other facets of the human-centered virtual machine, which have not been specifically looked at until now. These are the role of computer-based intermediaries (for back up support in problem solving), distributed communication in human-centered multi-agent systems and using computer-based artifacts to model breakdowns in human decision making. The role of intermediaries and distributed communication is illustrated by way of an intelligent distributed multi-agent multimedia euchre application on the net. In addition, aspects of a sales recruitment and benchmarking application are described to illustrate how computer-based artifact is used to overcome breakdowns in the sales, customer service and telesales recruitment decision making. In fact the modeling of the breakdowns is the prime motivator for the sales managers and sales personnel to use the computerized system for management of recruitment and benchmarking. The sales recruitment system described in this chapter (and developed by the first author) has been used in the industry for approximately the past three years.

In the last three chapters we have described applications of HCVM in medical informatics, image processing, internet games and sales recruitment. In these applications we have looked at multimedia artifacts in terms of how they can be used for improving the representational efficiency, effectiveness and interpretation of computer-based artifacts and also to some extent how they can be used for perceptual problem solving. However, there are interesting research issues and problems associated with management and retrieval of text, image, video, audio artifacts from single and multimedia databases. The area of multimedia information management and retrieval has in recent times become important due to the advent of the internet and World Wide Web (WWW). The solution to a number of issues in this area today lies with traditional solutions used for conventional databases. Recently, intelligent agents and human-centered information retrieval models are being proposed by researchers to address the problems in this area. In order to gain a proper understanding of the various problems and issues involved, we start this chapter by outlining the three evolution stages of this time honored area. We then describe these evolution stages in some detail. We include in this description applications and solutions proposed for multimedia information management and retrieval. We conclude the chapter by outlining open research problems in this area.

The objective of this chapter is to show the application of the conceptual and computational facets of the HCVM in a clinical drug prescription monitoring activity. The application is in the area of infectious diseases (Gorbach et al. 1998; Barrows et al. 1991) and addresses problems related to the gathering of patient symptomatic data, providing diagnostic assistance, and finding inconsistencies in practitioner prescribed treatments compared to those recommended by therapeutic guidelines (TG 1998).

In the last two decades digitial revolution has invaded business enterprises. Computers have enabled organizations to store gigabytes of data related to stock markets, electricity consumption profiles, troubleshooting and diagnostic data, etc.. As outlined by Fayyad and Uthuruswamy (1996a), in scientific endeavors, data represents observations carefully collected about some phenomenon under study. In business, data captures information about critical markets, competitors, and customers. On the other hand, in manufacturing, data captures performance and optimization opportunities, as well as the keys to improving process and troubleshooting problems. The reason organizations store or collect all this data is to enable them to extract some useful knowledge (at a later date!!) which can make them more productive, efficient, and competitive. The terms, Knowledge Discovery in Databases (KDD), and Data Mining have emerged in the last five years from this need of extracting useful knowledge. KDD is a nontrivial process of identifying valid, novel, potentially useful, and ultimately understandable knowledge from data (Fayyad et al. 1996b). Data mining is a step in the KDD process that consists of applying data analysis and discovery (learning) algorithms that produce a particular enumeration of patterns (or models) over the data. In fact, the data mining process can fit into the framework of intelligent transformation systems described in chapter 3. It will be noticed in this chapter that the transformation process is largely a bottom up knowledge engineering strategy.

In this chapter we outline a generic neuro-expert system (GENUES) architecture for hybrid reasoning. The architecture consists of five phases, namely, decomposition phase, control phase, decision phase, preprocessing phase, and postprocessing phase. The architecture is particularly applicable in time critical diagnostic/classification domains and data intensive domains in general. We describe the application of GENUES in a real time alarm processing system in a power system control centre.

In the last decade intelligent methodologies like artificial neural networks, fuzzy systems, genetic algorithms, and knowledge based systems (or expert systems) have been applied in a number of industrial applications. However, the approach of using these technologies by themselves alone has exposed some limitations which have discouraged their use in some industrial applications. For example, knowledge based systems are also sometimes known as “sudden death systems” because of their brittleness, and lack of graceful degradation. Artificial neural networks on the other hand, have problems generating explanations for their results (although, to some extent this problem has been addressed by the rule extraction techniques described in the last chapter). Genetic algorithms and fuzzy systems are known to be computationally expensive. Besides, as indicated in the last chapter, fuzzy systems have problems in determining the membership functions.

Real time applications to control industrial, medical, scientific, consumer, environmental and other processes is rapidly growing. Today such systems can be found in nuclear power stations, computer-controlled chemical plants, flight control, etc.. This growth however, has also brought to the forefront some of problems with the existing technologies. These problems have pushed for research into new techniques which could be used for solving these problems. 1991; Liu et al. 1992). The problems range from the enormous size of the power system to the fast response time constraints in emergency situations. This is the primary reason for selecting this domain as the domain of application.

An architecture is an information processing mechanism that can operate on the information represented in a form that is specific to the architecture (Chandrasekaran 1990). In the last chapter the TSL associative hybrid architecture has been described. In this chapter the primary goal is to outline the framework of the computational level architecture. This is done through the integration of the object-oriented model, agent model, and distributed operating system process model with the TSL architecture. Through this integration, an Intelligent Multi-Agent Hybrid Distributed Architecture (IMAHDA) with object-oriented properties is realized. IMAHDA at the computational level is built on four layers, namely object layer, software agent layer, intelligent agent layer, and problem solving agent layer. The PAGE descriptions of the software agent, intelligent agent, and problem solving agent layers are also outlined.

The hybrid architectures described in chapters 3, 4, and 5 are based on fusion, transformation, and combination of intelligent methodologies like expert systems, fuzzy logic, neural networks, and genetic algorithms. The concepts of fusion, transformation, and combination have been used in different situations or tasks, and by applying top-down and/or bottom-up knowledge engineering strategy. All these hybrid architectures have a number of advantages in that the hybrid arrangement is able to successfully accomplish tasks in various situations. However, these hybrid architectures also suffer from some drawbacks. These drawbacks can be explained in terms of the quality of solution and range of tasks covered (see Figure 6.1). Fusion, and transformation architectures on their own do not capture all aspects of human cognition related to problem solving. For example, fusion architectures result in conversion of explicit knowledge into implicit knowledge, and thus loose on the declarative aspects of problem solving. Thus, they are restricted in terms of the range of tasks covered by them. The transformation architectures with bottom up strategy get into problems with increasing task complexity. Therefore the quality of solution suffers when there is heavy overlap between variables, where the rules are very complicated, the quality of data is poor, or data is noisy. Also, because they lack explicit reasoning, the range of tasks covered by them becomes restricted. The combination architectures cover a range of tasks because of their inherent flexibility in terms of selection of two or more intelligent methodologies.

The need for intelligent systems has grown in the past decade because of the increasing demand on humans and machines for better performance. One of the reasons for this increasing demand is that we are passing through an era of information explosion, information globalization and consequently ever increasing competition. The information explosion has put huge constraints on time in which decisions have to be made. Today knowledge has become an important strategic resource to help humans deal with the complexity and sheer quantum of information. This strategic resource will help them to enhance their own performance and that of the machines they work with. The areas like knowledge discovery and data mining and soft computing are the latest manifestations of intelligent systems and knowledge in general.

In this paper we propose an integrated model which is derived from the combination of a generic neuro-expert system model, an object model, and unix operating system process (UOSP) model. This integrated model reflects the strengths of both artificial neural nets (ANNs) and expert systems (ESs). A formalism of ES object, ANN object, UOSP object, and problem domain object is used for developing a set of generic data structures and methods. These generic data structures and methods help us to build heterogeneous ES-ANN objects with uniform communication interface. The integrated model is applied in a real-time alarm processing system for a non-trivial terminal power station. It is shown how features like hierarchical/distributed ES/ANN objects, inter process communication, and fast concurrent execution help to cope with real-time system constraints like, continuity, data variability, and fast response time.© (1993) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

This work approaches the problem of modelling interactions between users and an adaptive system. It is presented the approach adopted in the MAIS project for modelling and designing a recommendation environment for delivering services in a personalized way. The model relies on semantic description of services, on rule-based user profile (part of these studies came from the experience in the KIWI project) and on the use of semantic matching algorithms. We also present the design of the recommendation engine, that is the core subcomponent of the Environment, which performs the task to evaluate each service according to user preferences.