Blackboard Architecture

This paper describes the Sentence Planner (sP) in the HealthDoc project, which is concerned with the production of customized patienteducation material from a source encoded in terms of plans. The task of the sP is to transform selected, not necessarily consecutive, plans (which may vary in detail, from text plans specifying only content and discourse organization to fine-grained but incohesive, sentence plans) into completely specified specifications for the surface generator. The paper identifies the sentence planning tasks, which are highly interdependent and partially parallel, and argues, in accordance with [Nirenburg et al., 1989], that' a blackboard architecture with several independent modules is most suitable to deal with them. The architecture is presented, and the interaction of the sentence planning modules within this architecture is shown. The first implementation of the sP is discussed; examples illustrate the planning process in action.

The paper presents a software re-engineering approach that is being used by a consortium of eleven academic institutions facilitated by a software engineering group to re-engineer previously developed natural language processing modules in laboratory environment, for developing eighteen machine translation (MT) systems covering nine pairs of Indic languages. This approach gives field deployable and maintainable MT systems (and modules) that can be reused as customizable components. The paper discusses its symbiotic approach of re-engineering modules, and also the experiences of developing MT systems in consortium mode. Following this approach, already seven MT systems have been developed and deployed on the web.

Abstra t. String mat h pro edures with respe t to two sets are investigated. The pro edures traditionally used for data ompression are based on singlestring mat h with respe t to a single set [LZ78, W84℄. Some re ent work broadened this view by presenting pro edures for multiple-string mat h with respe t to a single set [FPC98, PFP99℄ with improved performan e as ompared to the single-mat h versions. In this work an algorithm based on double-mat h with respe t to two sets is stated. We do onje ture that multiple-string mat h pro edures with respe t to two sets an a hieve even better performan e. A preliminary analysis orroborating this onje ture with some eviden e is reported in this work.

In this paper we address the problem of how new categories and new interpretations can be created in legal reasoning. We present some examples of such creativity, and then analyze them to identify the mechanisms that must be modeled for incorporating creativity. We point out that many of these mechanisms can be implemented by using components of already existing systems. Based on this analysis, we outline our approach, which uses a blackboard style architecture to generate creative arguments. Top-down processes are activated by portions of rules or ratio decidendi of precedents. Bottom-up processes are activated by the facts of the current case and precedents. The retrieval of precedents is also modeled as a blackboard process, so that which precedents are examined is determined dynamically depending on the contents of the blackboard. We compare our approach with existing research and then briefly mention future research problems.

Hospital laboratories perform examination tests upon patients, in order to assist medical diagnosis or therapy progress. Planning and scheduling patient requests for examination tests is a complicated problem because it concerns both minimization of patient stay in hospital and maximization of laboratory resources utilization. In the present paper, we propose an integrated patient-wise planning and scheduling system which supports the dynamic and continual nature of the problem. The proposed combination of multiagent and blackboard architecture allows the dynamic creation of agents that share a set of knowledge sources and a knowledge base to service patient test requests.

This paper describes a distributed dialogue management scheme for speechbased information seeking dialogue. The dialogue management is distributed to several components, supported by a general blackboard-type architecture for speech systems. By breaking down the dialogue management, we can achieve more general solutions and support sophisticated decision making algorithms.

Containers are currently used in the transport of products because of the point-to-point service they offer. By using specialised cranes, operations in maritime transport and land transport can be achieve quickly. To optimise operation costs, the whole management of the terminal is computerised proportionally with the complexity of the problem. This paper studies the problem from the point of view of the integration of different technologies supporting communication, data management, control, and human interface. This integration of heterogeneous intelligent components is made possible by means of a communication system that is the base of an object oriented distributed system. Using this system, "operation agents" that gain access to information in a high level and transparent way can be developed.

SHRUTI is a model of how predicate relations can be represented and reasoned upon using a network of spiking neurons, attempting to model the brain’s ability to perform reasoning using as biologically plausible a means as possible. This paper extends the biological plausibility of the SHRUTI model by presenting a genotype representation of connections in a SHRUTI network using indirect encoding and showing that working networks represented in this way can be produced through an evolutionary process. A multi-objective algorithm is used to minimise the error and the number of weight changes that take place as a network learns.

We discuss several aspects of legal arguments, primarily arguments about the meaning of statutes. First, we discuss how the requirements of argument guide the specification and selection of supporting cases and how an existing case base influences argument formation. Second, we present ,our evolving taxonomy of patterns of actual legal argument. This taxonomy builds upon our much earlier work on 'argument moves' and also on our more recent analysis of how cases are used to support arguments for the interpretation of legal statutes. Third, we show how the theory of argument used by CABARET, a hybrid case-based/rule-based reasoner, can support many of the argument patterns in our taxonomy.

passive surveillance systems that use waveform analysis as their primary means of detecting, classifying and locating a specific target. Discussion is further limited to those passive surveillance systems which must deal with considerable noise in the data. Present methods, which use visual examination of the waveform data for the detection of target waveforms, is complicated, time consuming, and requires considerable expertise. The lack of prior knowledge of the nature of the noise, (e.g., frequency spectra, amplitude, or dynamics), means that the majority of signal analysis must be done by experts. This study discusses and recommends a rule-based system which uses the following artificial intelligence structures: the blackboard architecture, and the frames data structure. Sources of uncertainty are also discussed and methods of dealing with it are suggested. This study recommends that the symbolic representation language be carefully selected for conciseness, efficiency, and a vocabulary rich enough to express everything desired by the experts. A learning knowledge source is also recommended.

The growing capabilities of information gathering systems in the submarine combat environment continues to generate interest in real-time decision support modules [N921, [P921, [S92], [V92]. The Maneuver Decision Aid (MDA) is a knowledge based expert system designed to simulate the command level human maneuver generation process. A key aspect of this model is that all identified situation goals and constraints are represented and combined according to their varying degrees of importance. This paper describes how constraint ratings are evaluated and combined within an object oriented blackbwrdframework.

Goal relationships resulting from the initial data and subsequent processing can he used to dynamically construct a partial topology of the solution space based on what appear to be feasible solutions. This structure can be used to make control decisions that significantly reduce the amount of search required to solve a problem in a complex domain. We examine the utility of this approach in the context of a multi-level, cooperative knowledge source model of problem solving. We present a taxonomy of goal relationships for ...

The goal of this research is to integrate aspects of artificial neural networks (ANNs) with symbolic machine learning methods in a probabilistic reasoning framework. Improved understanding of the semantics of neural nets supports principled integration efforts between seminumerical (so-called "subsymbolic") and symbolic intelligent systems. My dissertation focuses on learning of spatiotemporal (ST) sequences. In recent work, I have investigated architectures for modeling of ST sequences, and dualities between Bayesian networks and ANNs that expose their probabilistic and information theoretic foundations. In addition, I am developing algorithms for automated construction of Bayesian networks (and hybrid models); metrics for comparison of Bayesian networks across architectures; and a quantitative theory of feature construction (in the spirit of the PAC formalism from computational learning theory) for this learning environment. (Haussler 1988) Such methods for pattern prediction will be useful for building advanced knowledge based systems, with diagnostic applications such as intelligent monitoring tools.

The theory of behaviour-based systems (or autonomous agents) constitutes a useful approach for the modelling of intracellular signalling networks. In this sense, a cell can be seen as an adaptive autonomous agent or as a society of such agents, where each can exhibit a particular behaviour depending on its cognitive capabilities. We present an intracellular signalling model obtained by integrating several computational techniques into an agent-based paradigm. Cellulat, the model, takes into account two essential aspects of the intracellular signalling networks: (1) cognitive capacities, which are modelled as the agent abilities to interact with the surrounding medium and (2) a spatial organisation, this last obtained using a shared data structure through which the agents communicate between them. We propose a methodology for the modelling of intracellular signalling pathway using Cellulat and we discuss the goal of a virtual laboratory based on our model and presently under development.

Control strategies in most compla p&lem-s0lving systems, though highly parameter&d, are not adaptive to the characteristics of the particular task being solved. If the characteristics of the task are atypical, a fiied control strategy may cause incorrect or inefficient mg.

This article presents a descriptive comparative study on two methods of teaching, the traditional and blended learning pedagogical approach in a Business Communication course. Forty-four (44) students from the two controlled groups were enrolled in the course for the first semester of school year 2017-2018 participated in the study. The findings indicated that respondents' performance in ENGL118 (English Composition) had an effect on their performance in BUS261(Business Communication) and there was a significant positive relationship between year level and final grade in BUS261. Therefore, it was concluded that the year level and grade in ENGL118 have correlationson student'sperformance in BUS261. The results of the assessments of the two groups were compared and indicated a significant difference in the results which was influenced by respondents' year level and mode of conducting assessments. It is further concluded, that if the characteristics of both groups were the same, blended would have been more effective than the traditional method.This study recommends adopting the blended pedagogical approach not only in BUS261 course but also to other courses as applicable.

We describe a biologically inspired memory in a multi-agent based robotic architecture. In this approach, memory and pattern recognition are intertwined to form a cognitive memory that is used for recognition of objects in a robotics environment. This memory is implemented in a multiple agent behavior based blackboard architecture as an object recognition agent. The agent performance is tested against a standard dataset with satisfactory results. The system is currently installed in a mobile robotic platform where its capabilities and applications are explored.

This paper presents MESSIE, a multi-specialist architecture for scene interpretation in a robotic application. MESSIE is a centralized hierarchical blackboard architecture. The generic model of objects and the explicit description of sensors and materials allow the use of an application independent interpretation strategy. Two remote sensing applications on 2D scene interpretation and a third one, presented in this paper, on 3D scene interpretation allow the validation of the proposed architecture as well as the main features of MESSIE. After a brief overview of the state of art in 3D object modeling and scene interpretation, we discuss the scene interpretation problem from the knowledge representation view point. Then the architecture of MESSIE, the object modeling and the processing strategies (object detection and scene interpretation) are described. Further an application of 3D indoor scene interpretation in mobile robot context is given. We also present an interpretation running example using constrained low-level feature extraction mechanism to improve the image segmentation results.

In the building block style layout, two-dimensional routers are often necessary. A switchbox router utilizing two layers i p presented in this paper. The routing technique we have implemented can roughly be described as follows. Each net is first partitioned into a set of subnets. This decompo4ition is driven by the minimum Steiner tree. The construction is then carried out on a step by step basis, which allows the system to dynamically take into account new information on the problem. This information is examined by a set of expert $)stems which decide which subnet \hould be routed next. This part has been implemented \\ith a blackboard architecture. The router can handle pins that are not on a grid on one set of parallel edges. In order to always protide a solution to a given problem, the router expands the switchbox by adding rows d,namicallj where the) are needed. This point will be justified. The program has been implemented in the Lisp programming language. Man) test cases have been run successfully and some results, including Burstein's difficult switchbox, are presented.

In this work we discuss the development of a web application in the domain of movie chains using a Blackboard architecture, which is a well-established style for solving the problem of control, communication and collaboration in a system; it has traditionally been accepted as adequate for heuristic problem solving though not generally used for web applications. We present and discuss how the Blackboard architecture is used as the style for a graduate course project in Software Engineering. Issues about the implementation of the architecture are described and an assessment using some software metrics is presented.

Mining the growing data issued from the interpretation of remotely sensed images to obtain the necessary information for land cover change studies becomes more difficult and makes the data volume problem particularly acute. Mitigating this problem requires using data efficiently as metadata for mining and selecting appropriate data for change studies. In this paper, we propose an integrate hierarchical approach based on the use of a blackboard architecture and multi-agent system and having a reasoning ability to find the best strategy to extract and create metadata about extracted objects. This architecture models relationship between objects and primitives extracted from images as metadata and use a transition diagram to handle temporal dependencies and perform the detection of temporal changes of objects. We validate our approach on a set of multi-temporal Spot images, to model the evolution of detected object Keywords-component: interpretation of remotely sensed images; hierarchical blackboard architecture; multi-agent system. I.

Much w ork has been done in dialogue modeling for Human-Computer Interaction. Problems arise in situations where disambiguation of highly ambiguous data base output is necessary. W e propose to model the task rather than the dialogue itself. Furthermore, we propose underspeci ed representations to represent relevant data and to serve as a base for generating clari cation questions that guide the user e ciently to arrive at his communicative goal. In this paper, we establish a connection between underspeci ed representations as representations of disjunctions and clari cation questions. Our approach to clarifying dialogues di ers from other approaches in that the form of the clari cation dialogues is entirely determined by the domain modeling and by the underspeci ed representations.

This paper investigates the effects of parallelism on blackboard system scheduling. A parallel blackboard system is described that allows multiple knowledge source instantiations to execute in parallel using a shared-memory blackboard approach. New classes of control knowledge are defined that use information about the relationships between system goals to schedule tasks-this control knowledge is implemented in the DVMT application on a Sequent multiprocessor using BBl-style control heuristics. The usefulness of the heuristics is examined by comparing the effectiveness of problem-solving with and without the heuristics (as a group and individually). Problem solving with the new control knowledge results in increased processor utilization and decreased total execution time.

Internet has succeeded as a global information system mainly because of its availability and openness, and the simplicity of its standards and protocols. However, the current use of Internet as universal middleware has clearly shown the lack of maturity of Web technology to support distributed applications, which involve communication, cooperation, and coordination. This paper proposes an architectural solution to solve these interaction restrictions. It is based on an extension of the service-oriented architectures, adding a new coordinator role that allows more flexible relationships between service providers and requestors than the provided by the client/server model. This role is inspired by the Blackboard architectural pattern and it is the conceptual basis of a Web-Coordination service able to coordinate distributed and heterogeneous applications through Internet. To prove the effectiveness of this proposal, the Web-Coordination service has been used in an highly dynamic and collaborative application context, the Location-Based Services.

Ⅰ. INTRODUCTION Current WWW, which is based on the HTML, is operated by browsing Internet information in user side and executing the program in server side. But Web Services, which is distributed component model, can access/use/reuse a remote web object by XML based standard protocol. Web Services is usually used to extending the functions of business domain applications. According to the characteristics of Web Services, it could support the diversity of terminal, network and user environment for ubiquitous computing environment. Ubiquitous computing network comprises a variety of distributed service devices. Today Web Services technology enables the heterogeneous devices to provide their own services and interact with each other via well-defined Internet protocol. Nevertheless, service devices in ubiquitous environment require more event-driven, autonomous interactions beyond the rather passive serviceoriented architecture of the present time. So, in this paper, the technologies of Web Services for ubiquitous environment is named as Ubiquitous Web Services(UWS). UWS shall support not only simple wired computing environment but also various terminals and computing environment (e.g. home network, Telematics, Broadband Convergence Network, Mobile, Broadcasting, etc.). It will be continuously enhanced and converged in multiple domains. Also, Ubiquitous Web Services technologies that support platform independent application integration cope with limitation of legacy middleware technologies based on distributed object. Therefore it could be software infrastructure for ubiquitous computing with other Web technologies. In this paper, we propose the development of ECA (Event-Condition-Action) rule description language for the ubiquitous environments in an attempt to support capability for autonomous interactions among service-oriented devices in ubiquitous computing network. The rules are embedded in distributed devices which invoke appropriate services in the network if the rules are triggered by some internal or external events. The presented ECA-based device coordination approach is expected to facilitate seamless interoperation among the Web Services-enabled devices in the emerging ubiquitous computing environment. Ⅱ. ECA RULE DESCRIPTION LANGUAGE Ubiquitous computing environments are becoming more heterogeneous and service rich domains[1]. Devices with particular services are interconnected each other via various types of networks. While web services technology has become a de facto standard for business application integration[2][3], it is also rapidly emerging as an effective means for achieving inter-operability in ubiquitous computing environments[4][5]. Also, in the current Web Services, the characteristics are described in WSDL, a services developer shall know the location of the WSDL, or find the appropriate WSDL in the service registry such as UDDI. It means that user can not know the additional and dynamic information like service policy, capabilities, state information, and etc. In addition, in the ubiquitous computing environment, a service shall add or remove momentarily. So, there is need to develop the specification for (i)dynamic service discovery, (ii)dynamic service binding, and (iii)dynamic service composition for the ubiquitous environments. An ECA (Event-Condition-Action) rule description language is introduced for event-driven coordination of Web Servicesenabled devices. The ECA rule is originally devised to provide traditional database systems with the capability of event-driven, 1/5

The theory of behaviour-based systems (or autonomous agents) constitutes a useful approach for the modelling of intracellular signalling networks. In this sense, a cell can be seen as an adaptive autonomous agent or as a society of such agents, where each can exhibit a particular behaviour depending on its cognitive capabilities. We present an intracellular signalling model obtained by integrating several computational techniques into an agent-based paradigm. Cellulat, the model, takes into account two essential aspects of the intracellular signalling networks: (1) cognitive capacities, which are modelled as the agent abilities to interact with the surrounding medium and (2) a spatial organisation, this last obtained using a shared data structure through which the agents communicate between them. We propose a methodology for the modelling of intracellular signalling pathway using Cellulat and we discuss the goal of a virtual laboratory based on our model and presently under development.

We describe a biologically inspired memory in a multi-agent based robotic architecture. In this approach, memory and pattern recognition are intertwined to form a cognitive memory that is used for recognition of objects in a robotics environment. This memory is implemented in a multiple agent behavior based blackboard architecture as an object recognition agent. The agent performance is tested against a standard dataset with satisfactory results. The system is currently installed in a mobile robotic platform where its capabilities and applications are explored.

Internet has succeeded as a global information system mainly because of its availability and openness, and the simplicity of its standards and protocols. However, the current use of Internet as universal middleware has clearly shown the lack of maturity of Web technology to support distributed applications, which involve communication, cooperation, and coordination. This paper proposes an architectural solution to solve these interaction restrictions. It is based on an extension of the service-oriented architectures, adding a new coordinator role that allows more flexible relationships between service providers and requestors than the provided by the client/server model. This role is inspired by the Blackboard architectural pattern and it is the conceptual basis of a Web-Coordination service able to coordinate distributed and heterogeneous applications through Internet. To prove the effectiveness of this proposal, the Web-Coordination service has been used in an highly dynamic and collaborative application context, the Location-Based Services.

Ⅰ. INTRODUCTION Current WWW, which is based on the HTML, is operated by browsing Internet information in user side and executing the program in server side. But Web Services, which is distributed component model, can access/use/reuse a remote web object by XML based standard protocol. Web Services is usually used to extending the functions of business domain applications. According to the characteristics of Web Services, it could support the diversity of terminal, network and user environment for ubiquitous computing environment. Ubiquitous computing network comprises a variety of distributed service devices. Today Web Services technology enables the heterogeneous devices to provide their own services and interact with each other via well-defined Internet protocol. Nevertheless, service devices in ubiquitous environment require more event-driven, autonomous interactions beyond the rather passive serviceoriented architecture of the present time. So, in this paper, the technologies of Web Services for ubiquitous environment is named as Ubiquitous Web Services(UWS). UWS shall support not only simple wired computing environment but also various terminals and computing environment (e.g. home network, Telematics, Broadband Convergence Network, Mobile, Broadcasting, etc.). It will be continuously enhanced and converged in multiple domains. Also, Ubiquitous Web Services technologies that support platform independent application integration cope with limitation of legacy middleware technologies based on distributed object. Therefore it could be software infrastructure for ubiquitous computing with other Web technologies. In this paper, we propose the development of ECA (Event-Condition-Action) rule description language for the ubiquitous environments in an attempt to support capability for autonomous interactions among service-oriented devices in ubiquitous computing network. The rules are embedded in distributed devices which invoke appropriate services in the network if the rules are triggered by some internal or external events. The presented ECA-based device coordination approach is expected to facilitate seamless interoperation among the Web Services-enabled devices in the emerging ubiquitous computing environment. Ⅱ. ECA RULE DESCRIPTION LANGUAGE Ubiquitous computing environments are becoming more heterogeneous and service rich domains[1]. Devices with particular services are interconnected each other via various types of networks. While web services technology has become a de facto standard for business application integration[2][3], it is also rapidly emerging as an effective means for achieving inter-operability in ubiquitous computing environments[4][5]. Also, in the current Web Services, the characteristics are described in WSDL, a services developer shall know the location of the WSDL, or find the appropriate WSDL in the service registry such as UDDI. It means that user can not know the additional and dynamic information like service policy, capabilities, state information, and etc. In addition, in the ubiquitous computing environment, a service shall add or remove momentarily. So, there is need to develop the specification for (i)dynamic service discovery, (ii)dynamic service binding, and (iii)dynamic service composition for the ubiquitous environments. An ECA (Event-Condition-Action) rule description language is introduced for event-driven coordination of Web Servicesenabled devices. The ECA rule is originally devised to provide traditional database systems with the capability of event-driven, 1/5

Research into multi-agent systems has increased over several years, leading to demand for multi-agent system simulators to investigate the interaction between different agents and their environment. The more complex and intelligent the agents become, the more difficult it is to simulate them. Distributing the simulation across different processors would improve its performance. This paper describes the implementation of a multi-agent test-bed called TileWorld on a distributed blackboard system, DARBS. The slowdown of the simulator on a single processor has been measured as the number of agents increases, and the likely effect of migrating to a multiple-processor system is assessed.

Conference calls represent a natural but limited communication channel between people. Lack of visual contact and limited bandwidth impoverish social cues people typically use to moderate their behavior. This paper presents a system capable of providing timely aural feedback enabling meeting participants to check themselves. The system is able to sense and recognize problems, reason about them, and make decisions on how and when to provide feedback based on an interaction policy. While a hand-crafted policy based on expert insight can be used, it is non-optimal and can be brittle. Instead, we use reinforcement learning to build a system that can adapt to users by interacting with them. To evaluate the system, we first conduct a user study and demonstrate its utility in getting meeting participants to contribute more equally. We then validate the adaptive feedback policy by demonstrating the agent's ability to adapt its action choices to different types of users.

The theory of behaviour-based systems (or autonomous agents) constitutes a useful approach for the modelling of intracellular signalling networks. In this sense, a cell can be seen as an adaptive autonomous agent or as a society of such agents, where each can exhibit a particular behaviour depending on its cognitive capabilities. We present an intracellular signalling model obtained by integrating several computational techniques into an agent-based paradigm. Cellulat, the model, takes into account two essential aspects of the intracellular signalling networks: (1) cognitive capacities, which are modelled as the agent abilities to interact with the surrounding medium and (2) a spatial organisation, this last obtained using a shared data structure through which the agents communicate between them. We propose a methodology for the modelling of intracellular signalling pathway using Cellulat and we discuss the goal of a virtual laboratory based on our model and presently under development.

This paper proposes to model the intracellular signalling networks using a fusion of behaviour-based systems and the blackboard architecture. In virtue of this fusion, the model developed by us, which has been named Cellulat, allows to take account two essential aspects of the intracellular signalling networks:(1) the cognitive capabilities of certain types of networks' components and (2) the high level of spatial organization of these networks. A simple example of modelling of Ca2+ signalling pathways using Cellulat is presented ...

This paper presents a model for the selection of optimal assembly sequences for a product in multirobot systems. The objective of the plan is the minimization of the total assembly time (makespan). To meet this objective, the model takes into account, in addition to the assembly times and resources for each task, the times needed to change tools in the robots, and the delays due to the transportation of intermediate subassemblies between different machines. An A* algorithm that solves the problem is also presented, which starts from the And/Or graph for the product (compressed representation of all feasible assembly plans).

This paper introduces CAMEO, a behavior-driven design approach to address commonly occurring technical and social problems in audio-only teleconference calls. Many of these problems are associated with the missing visual channel and the low bandwidth for non-verbal signals. CAMEO seeks not only to sense these problems, but also to frame and respond to them in considerate ways. These include scheduling of advisory feedback prompts, and assistive feedforward mechanisms to augment this bandwidth constrained medium. This paper describes their implementation in CAMEO using a blackboard architecture that shapes and define its behavior. Two experiments were conducted to evaluate CAMEO on its resolution of conversational dominance in a collaborative meeting, and its utility in reducing the effects of disruptive background noise on a conference call. The participants were asked to solve hangman and chess puzzles by collaborating on a multiparty conference call. We show that variance in conversational dominance can significantly be reduced with proactive aural feedback. Our experiments further reveal that such feedback can also reduce the impact of background noise on conversations.

We explore a goal-oriented as opposed to a problem-oriented approach to DSS development for ecosystem management. Ecosystem management ordinarily is guided by a set of goals that may conflict in various ways. Problems are perceived obstacles to realizing goals. Identifying and resolving conflicts between goals, testing current or projected situations for goal satisfaction, and problem identification all require a robust model of the goal structure for the intended domain. The lowest level of this goal structure must be represented as a'esirable jirture conditions consisting of proposed values for observable indicators. A model of the causal, legal, and other institutional relations between these desirable future conditions is also needed. Two projects based on a goal-oriented approach to DSS development are described. The first project has produced an initial prototype that incorporates goals for forest management in rules representing three tiers: management unit goals, stand-level goals, and desirable future conditions. The second, at an initial knowledge acquisition stage, is an attempt to develop a participatory decision-making methodology for socially and environmentally sensitive economic development in Central America.

The paper presents a software re-engineering approach that is being used by a consortium of eleven academic institutions facilitated by a software engineering group to re-engineer previously developed natural language processing modules in laboratory environment, for developing eighteen machine translation (MT) systems covering nine pairs of Indic languages. This approach gives field deployable and maintainable MT systems (and modules) that can be reused as customizable components. The paper discusses its symbiotic approach of re-engineering modules, and also the experiences of developing MT systems in consortium mode. Following this approach, already seven MT systems have been developed and deployed on the web.

With increased number of new services and users being added to the communication network, management of such networks becomes crucial to provide assured quality of service. Finding skilled managers is often a problem. To alleviate this problem and also to provide assistance to the available network managers, network management has to be automated. Many attempts have been made in this direction and it is a promising area of interest to researchers in both academia and industry. In this paper, a review of the management complexities in present day networks and artificial intelligence approaches to network management are presented. Published by Elsevier Science B.V.

As companies increasingly customize their products, move towards smaller lot production and experiment with more flexible customer/supplier arrangements, they increasingly require the ability to respond quickly, accurately and competitively to customer requests for bids on new products and efficiently work out supplier/subcontractor arrangements for these new products. This in turn requires the ability to rapidly convert standard-based product specifications into process plans and quickly integrate new orders with their process plans into existing production schedules across the supply chain. This paper describes IP3S, a blackboard-based agent for supporting integrated process planning/production scheduling across the supply chain. IP3S agents support concurrent development and dynamic revision of integrated process-planning/productionscheduling solutions across the supply chain, maintenance of multiple problem instances and solutions across the supply chain, flexible user-oriented ...

As companies attempt to increase customization levels in their product offerings, move toward smaller lot production, and experi ment with more flexible customer/supplier arrangements such as those made possible by electronic data interchange (EDI), they increas ingly require the ability to (1) respond quickly, accurately, and competitively to customer requests for bids on new or modified products and (2) efficiently work out supplier/subcontractor arrangements for these products This in turn requires the ability to (1) rapidly convert standard-based product specifications into process plans and (2) quickly integrate process plans for new orders into the existing production schedule to best accommodate the current state of the manufacturing enterprise

The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.

Conference calls represent a natural but limited communication channel between people. Lack of visual contact and limited bandwidth impoverish social cues people typically use to moderate their behavior. This paper presents a system capable of providing timely aural feedback enabling meeting participants to check themselves. The system is able to sense and recognize problems, reason about them, and make decisions on how and when to provide feedback based on an interaction policy. While a hand-crafted policy based on expert insight can be used, it is non-optimal and can be brittle. Instead, we use reinforcement learning to build a system that can adapt to users by interacting with them. To evaluate the system, we first conduct a user study and demonstrate its utility in getting meeting participants to contribute more equally. We then validate the adaptive feedback policy by demonstrating the agent's ability to adapt its action choices to different types of users.

Syndromic surveillance requires the acquisition and analysis of data that may be suggestive of early epidemics in a community, long before there is categorical evidence of unusual infection. These data are often heterogenous and noisy, and need to be interpreted by a combination of analytic methods. Syndromic surveillance thus involves problems of integrating data, of configuring problem-solving strategies, and of mapping integrated data to appropriate methods. These tasks have been studied in the knowledge-based systems community for many years. We present a software architecture that suppports knoweldge-based data integration and problem solving, thereby facilitating many aspects of syndromic surveillance. Central to our approach, a set of reference ontologies supports semantic integration and a parallelizable blackboard architecture implements invocation of appropriate problem-solving methods and control of reasoning. We demonstrate our approach with BioSTORM-an experimental system that offers an end-to-end solution to syndromic surveillance.

This paper describes a recently developed maintenance system for a microsatellite, which is being constructed and it is expected to be launched shortly. The autonomous maintenance system is going to be installed on a personal computer located in an earth-based control room. As this place will most likely be sta ed 24 h a day, 7 days a week, there is a good deal of interest in the use of increased automation in maintenance tasks in order to improve the e ciency with which personnel are used and as a way to reduce costs. The system described here is a good example of emerging automation technology that is intended to replace human operators responsible for system maintenance. The structure of the automation system is based upon an architecture of collaborative intelligent agents designed to detect failure in any of the microsatellites components. The multiagent system consists of a set of di erent agents devoted to failure detection, prevention and correction. Regarding correction, speciÿc agents for each constitutive part of the microsatellite have been developed that take over the necessary actions to solve any given problem in its operation. The detection agent decides which correction agent control should be transferred, based upon inference obtained from its knowledge base made up of rules for testing and diagnosis. Actions for correction may imply the use of redundant systems, which can reconÿgure themselves to avoid defective circuits, among other repairing strategies. The prevention agent uses predictive models that have been developed for each signiÿcant failure mode. Statistical models are also used by this agent to determine the shape of the distribution of times to failure. The prevention agent selects the corresponding correction agent to which control is going to be transferred and this agent carries out the necessary actions to prevent the system failure. The overall intelligent system employs a blackboard architecture for communication and collaboration among agents. Several simulations with specially designed test cases used to evaluate the system performance suggest that in order for maintenance automation to be e ective, it must be designed in close relationship with human operators who will occasionally troubleshoot, maintain and repair the microsatellite from earth. This human-centered approach of the design implies assigning particular importance to the e ectiveness achieved in the process of knowledge acquisition when the intelligent agents are being designed.

Non-destructive evaluation is widely used in the manufacturing industry for the detection and characterisation of defects. Typical techniques include visual, magnetic particle, fluorescent dye penetrant, ultrasonic, and eddy current inspection. This paper presents a multi-agent approach to combining image data such as these for quality control. The use of distributed agents allows the speed benefits of parallel processing to be realised, facilitating increased levels of detection through the use of high resolution images. The integration of multi-sensor devices and the fusion of their multi-modal outputs has the potential to provide an increased level of certainty in defect detection and identification. It may also allow the detection and identification of defects that cannot be detected by an individual sensor. This would reduce uncertainty and provide a more complete picture of aesthetic and structural integrity than is possible from a single data source. A blackboard architecture, DARBS (Distributed Algorithmic and Rule-based Blackboard System), has been used to manage the processing and interpretation of image data. Rules and image processing routines are allocated to intelligent agents that communicate with each other via the blackboard, where the current understanding of the problem evolves. Specialist agents register image segments into a common coordinate system. An intensity-based algorithm eliminates the landmark extraction that would be required by feature-based registration techniques. Once registered, pixel-level data fusion is utilised so that both complementary and redundant data can be exploited. The modular nature of the blackboard architecture allows additional sensor data to be processed by the addition or removal of specialised agents.

The framework of an automated vision system for the monitoring of quality control is presented. Inspection which is capable of detecting various forms of defects is achieved by combining distributed artificial intelligence and image processing. The blackboard architecture DARBS (Nolle and Wong 2001) manages the processing of image data via an area of shared memory where the current understanding of the problem evolves. Registration into a common coordinate system and segmentation of reference and sensed images is performed by intelligent agents which communicate with each other by means of the blackboard. Pixel-level fusion is performed on registered images in order to exploit complementary and redundant data, allowing identification of suspected defects. The difficulties of landmark extraction common to feature-based registration techniques have been replaced by an intensity-based algorithm. Addition or removal of specialised agents is simplified by the blackboard's modular nature.