Home | Conferences | Techknowledge | 12th National Conference | Dynamic Access And Emergence Mechanism For Collaborative Adaptive Systems From Agent’s Base

Dynamic Access And Emergence Mechanism For Collaborative Adaptive Systems From Agent’s Base

By
Font size: Decrease font Enlarge font

1. Introduction: The main issue in human-machine interaction is obtained a collaboration situation between a human user and a computer system. The system must be attuned to the user, and the user to the system. Good conventions and guidelines shift the entire burden of adaptation to the user adoption, and the design restrictions that they impose are geared towards easing this task form the user to from specification and requirements of adoption by learning strategies [1].But providing such a list of technologies does not capture the essential feature of the intelligent interface research area: an intelligent interface must utilize technology to make an improvement: the resulting interface should be better than any other solution, not just different and technically more advanced but should be more capable on aspects of technology, eases of acceptance and flexible to move between environment: user modeling and natural language dialogue used to denote a scheme of  that the system maintains, and adapts its behavior to interface access adoption  by collaborative learning from system agent base and same time maintaining a system base for interface adoption between system to system environment and system to user access and process listing in selected services and grouping of protocol enabled agent base which maintain system environment access and learning base of adoption strategies. Development of adaptive systems reinforces the need to know thy system environment . The interface is set at design time but it is changeable for objective perspective. Adaptive systems may require more up-front environment analysis, since not only does the system designer need to know the user, that knowledge also must be embedded into the system [2].

Moreover, the system needs to known when and how to act on that knowledge gained from system base and learning base of adoption using HCI principle adoption for environment access between interface [3]. Evaluate the risks, costs and benefits to make sure the payoff is worth the implementation effort for their specific circumstances. Organizations that can amortize the cost over large number of agent’s base may find it easier to justify such a system and supporting system may need to be revised to handle adaptive systems in generic adoption and learning on feed sequences of agents.

2. Intelligent Interface type Modeling and Adoption Strategies of Agents: The area of Intelligent Interfaces is one of the most heterogeneous research subjects dealing with computers that exist. In this area, people from vastly different disciplines and research areas within disciplines meet, debate and collaborate. The term is so wide that people will shrink from it in practice Counts as an intelligent interface adoption technique will vary over time, but the following list is a fairly complete list of the kinds of techniques that today are being employed in intelligent interfaces [4], user adaptive, modeling, and natural lingual [6] interpreter and dialogue enabled using learning and explanation generator .A system state that maintains a model may be adaptable or self-adaptive. The distinction can be made more fine-grained; [4] distinguish between several levels of adaptively, depending on who takes initiative to the adaptation, who proposes the adaptation, who decides upon it, and who carries it out. System based on natural language dialogue is directly inspired by the thought of getting a computer to carry out a human-like dialogue. Since people are able to interact with each other in natural language, it should be natural and easy to interact with a computer in the same manner. The research has many facets, ranging from the literal interpretation of natural language sentences to recognizing the focus and topic shifts of natural dialogue [2, 5].  Find such research on text processing that is necessary to enable advanced information filtering. An important area of research for intelligent interfaces is to integrate several ways of interaction in a multimodal human - computer interaction [5]. In interaction, natural language is a central and important ingredient, but it is not a target goal of its own, and deficiencies in language understanding can be compensated by the ability to interact using other modalities such as direct manipulation and interface access and consideration. These are very ambitious goals. Some of the early models of human cognition in interacting with computer interfaces aimed to be analytical in this sense [7]. Agent base learning can be used to estimate the cognitive load on agents in routine interface tasks analysis at a rather low level of detail, and provide little insight in what is an appropriate design of an interface and extend and modify them to be applicable to the new functionalities and interaction principles found in intelligent interfaces [8]. The prevailing development strategy is that of iterative and user-oriented design, where the interface is repeatedly tested with agents to refine the design, and see which adaptations work and which do not work. If an interface is self-adaptive, the same applies. A user/agent must be able to inspect why a certain adaptation was generated, and correct the behavior if the result was not what the user wanted [8]. Inspection is also important to allow the user to trust a system: an expert agent base must be able to produce an explanation of certain action. Else, a user/agent may ignore the system's advice because it mistrusts its competence. Intelligent interfaces may provide both active and passive adaptations to the user's needs. These require different interaction metaphors to be understood by the user. The prevailing interaction metaphor is that of direct manipulation [10]. The system must behave rather passively, and let the user maintain the initiative and control of the interaction. The intelligence of the system may show only in the set of options that the system suggests.[9]. The adaptation process supports different types of adaptations based on each partial profile. We begin with a general description of the elements included in this process, continuing with the two main tasks that are the base of our proposal for the personalization process to be done in a Collaborative adaptive system: The Agent learning modeling task and the adaptation task. The first one covers three main objectives[11,12]: 1), define the agent’s characteristics that are relevant into a specific context to generate useful   base and an adaptations 2), define how these characteristics can be obtained or inferred and 3), define how these characteristics contribute to the adaptation process. The second one describes the adaptation mechanism itself, that is, how the users’ characteristics are managed to generate an adjusted learning design path. Collaborative adaptive system, combines different types of adaptation in order to cover more completely the heterogeneity of users. This complete adaptation   consist [12,13] in 1), delivering the learning objects adjusted to the specific competence level of the agents base 2), delivering ranked learning objects according to the agent  learning style; 3), suggesting future actions to agents  according to their collaborative behavior; 4), guaranteeing access to the learning objects according to the agent and environment  characteristics.
3. Agents Base for Interface Access Monitoring Methods between Environments: Era of research into the design and engineering principles necessary to build and manage system, models and design principles that allow for the creation of emergent behaviors; tools and methods that support such development; and principled approaches to incremental development starting  filtering techniques are applied[15]. Some Recommendation Agents [11] are the following:
•    Collaboration agent: in charge of providing recommendations to promote the collaboration among the activity and member interface and modules and sub modules [12].
•     Pedagogical agent: in charge of suggesting activities and learning objects which are not mandatory according to the specification in early stages but can be relevant for the learner in the current situation[13]. (e.g. lack of knowledge or high interest),
Presentation agent: in charge of selecting the content to be presented to the learner, taking into account device capabilities or accessibility requirements and adoption of interface according to need in scenario[20]. Each Recommendation Agent contacts the appropriate agent’s base to gather information about the learners, to interact with interface module handler, contents, device and interactions basis modeling .
3.1 System Environment Collaboration Learning: An objective of collaborative learning [19] for accessing environment can be identifying:

(a), A specific statement about what system environment is expected to learn or to be able to do as a result of collaborative accesses: more specifically this is a learning objective 
(b), A measurable operationalisation of a policy, strategy or mission: this is an implementation objective. An objective is a description of an intended outcome, written in specific terms (fig 2), It describes:
•    What does agent learning will produces.
•    By how much recourse it can be accessed and how much time it consumes.
•    Using what agents the interface is adoptable (application, equipment, facilities, services),
•    To what standard. (Reliability and feasibility),
In the following we specify how environment access rules are used at each level.
At learning time, a system is modeled using a base model which contains its basic functionalities and a set of variant models[16] which can be composed with this base model depending on the context of the accessing interface system, an adaptation model is specified. The adaptation model specifies the following elements:
•    A set of dependencies that specify constraints on variants and a context model of the system.
•    A set of adaptation rules that link the context elements from the context model to the variants that should be used according to the context variation.
•    Is interface agent is usable calendaring assistant?
•    Does interface agent increases task effectiveness?
•    Is interface agent a potentially useful assistant? (e.g., selective actions , ratings, usage frequency and action sequences),
Environment data and elements environment, and the user’s current location, for example Ease of adoption of the agents interface by systems will determine success or failure of a deployed evaluation strategy.

4. Agents Base Adaptive Strategies and Modeling of Environment Access: Endowing a system with adaptive learning property [17] can take many different shapes. The way self-adaptation or learning adoption  has to be conceived depends on various aspects, such as, system requirements, environment characteristics, agent failure ,interface assumption  and other system properties. Understanding the problem and selecting a suitable solution requires precise models for representing important aspects of the self-adaptive system, its systems and role of the interface for collaboration, and its environment. We provide a strategy to applying agents dimensions for self-adaptive systems. Each dimension describes a particular aspect of the system that is relevant for access using agent base and learning and selection.

4.1 Dimensions of Modeling Agent Base (Adaptation Process and Techniques),: Adaptation will be the first group describes the modeling dimensions related to adaptation based on identification of type of availability ,degree of automation, form of environment access and modifiability ,place of environment access and impact of access and adoption ,abstraction of environment and transparent access technique of decision making and degree of decision.
 


4.2  Performance of Agent Base: The second group describes modeling dimensions related to timing issues of agent access and generation from the base to an environment for time based access into environment and system can be identified using access time of agent into specific environment, and response time taken in learning and adoption and performance of learning and triggering of agent between system and environment.

4.3  Dependability of Agents Base and Access to System: The third and final group we consider describes modeling dimensions related to dependability, that is, the ability of a system to deliver a service that can justifiably be trusted. Reliability, availability, and confidentiality are attributes of dependability and access security. The domain of each of these properties ranges from high to low. In the Agents base adoption, the reliability of the interface avoiding a collision is expected to be high.

5.  Agent Base Learning Strategies Models: The goal of Agent’s assurance is simple.  Adoption of agent’s base need to provide evidence that the set of stated functional and nonfunctional properties are satisfied during system’s operations. While the goal is simple, achieving it is not. Traditional verification and validation methods [22], static or dynamic, rely of stable descriptions of environment access and system models and properties

 
The characteristics of self-adaptive systems create new challenges for developing high-assurance systems. Current verification and validation methods align well with changing goals and requirements as well as variable system functionality adaptive systems implied framework. The idea of the approach we propose is to combine learning and agent base techniques to handle the complexity of adaptive system construction and execution. System copes with complexity through abstracting the dynamic variability. Learning techniques ( fig 4), are utilized to model the adaptation concerns separately from the other aspects  abstractions and advanced separation of concerns, the adaptation becomes easier to design and understand, possible to validate and allows to easily evolve the adaptation policies  at   dynamic access.

5.1.  Learner Rules & Responsibilities: The agent base developer’s foremost responsibility is to ensure the compliance with the interface environment standard. But, complying with the technical environment is not the only task of the learner.

 
Agent’s base developer is also the expert who knows best how an existing system needs to be migrated and needs to decide (fig 5), if individual functional components or the entire model as such is migrated for parsimony reasons small functional units are preferred, but economic, commercial and practical reasons may lead to larger components (which internally can again be interface compliant),
 

Second, needs to decide which data need to be exposed to the outside world both from the viewpoint of providing and accepting values. This work will usually also involve the modeling learning base. Which data need to be exposed is very dependent on the use of the environment component. In many cases it will be sufficient to expose data that need to be exchanged by environment. But if scenarios need to be calculated it may be necessary that underlying schematizations can be altered using the agent and become accessible through the agent base interface. It remains the responsibility of the agent’s developers to decide if and how such data can be altered in an environment with agent’s base cope with access on interoperability. Though the environment learning and agent selection design allows all models to calculate at their own time stepping and spatial representation, the mechanism requires that output can be delivered at any given time and place (reasonably within the models overall temporal and spatial boundaries),[16] .To achieve this system  needs to know at what time step it is. Based on this knowledge the system component [fig 6] would do

•    Start calculations, if the required time step has not been reached but can be computed.
•    Extrapolate in time if the required time step cannot be reached.
•    Search its buffer (if implemented), if the requested time step has passed, otherwise calculate required values),
•    Interpolate between different time steps if the requested time step does not match the internal time stepping.
•     Aggregate in space and time if the requested spatial and temporal representation does not match the internal representations.

To support the agent developer the smart utility and learning base adoption will provide a library containing some basic functionality to support the actions listed in course of access. In advanced integrated systems iterations and optimizations may be required to facilitate efficient computing to access environment and learning base.
6. Future Work: The development of system in presented Schema is needs to be a improved quality for schema sharing and joint activity and it needs to make more share global agent base and learning is optimized for every consequence agent base access and interface access environment is need to capitalized more resources as it have and automated learning can be developed for mapping in between environment for reducing collision of shared activity and shared agent base up to a global efficiency to local efficiency.
7.  Conclusion: System access and adaptive techniques provide a way to optimize a user interface for individual users. These techniques can also be used to evolve an interface as the user becomes more skilled with a system agent base more empirical evaluation and practical experience will be needed to determine how useful they are and for what applications. The cooperation involves communication and a specific kind of coordination. From these elements result the development of a new paradigm of the collaborative activities. The development of system in presented schema is accelerated, along with the activity networks and, the quality characteristics become strictly related to the security characteristics. Collaborative work can be successful if agent base has proper diagnosis of situation and show goodwill and responsibility throughout the learning. The development of collaborative systems conduct to increase their complexity and the global character of the economy is designed to determine, also a global character for many of the collaborative systems. From the information point of view, to these global collaborative systems must correspond global performance indicators, procurement systems scratchy and data conversion procedures, to transform heterogeneous information into homogeneous entries for aggregate indicators, defined in the metrics of collaborative systems. Based on these aggregated indicators should decide appropriate to the global level, intermediate level and the execution level of any collaborative system organized into hierarchical levels.
 
References

1.    http://www.collaborative-systems.org
2.    types of interface  in adoption Annika Wærn march 1997available from   http://www.sics.se/~annika/ii_links.html
3.    HCI: Project Avanti [online]. Sankt Augustin, Germany, February 1999. Available from the World Wide Web: http://fit.gmd.de/hci/projects/avanti
4.    Uwe Malinowski, Thomas Kühme, Hartmut Dieterich and Matthias Schneider-Hufschmidt, (1992), A taxonomy of adaptive user interfaces. in ed. A. Monk, D Diaper and M D Harrison, HCI'92, People and Computers VII, Cambridge University Press.
5.    Barbara Grosz and Candace Sidner, (1986), Attention, intention and the structure of discourse. Computational Linguistics , pages 175-204.
6.    Ivan Bretan, (1995), Natural Language in Model World Interfaces. Licentiate thesis, Dept. of computing and system sciences, Stockholm University.
7.    R. W. Southwick, (1989), Explaining reasoning: an overview of explanations in knowledge based systems. The Knowledge Engineering Review, pages 1-19.
8.    Kristina Höök, (1997), Evaluating the Utility and Usability of an Adaptive Hypermedia System, in Proceedings of the International Conference on Intelligent User Interfaces, Orlando, Florida, January. ACM
9.    T. Kuhme, U. Malinowski and J.D. Foley, (1993), Adaptive Prompting. Technical report GIT-GVU-93-05, Georgia Institute of Technology.
10.    Pattie Maes, (1994), Agents that reduce work and information overload. Communications of the ACM 37(7),
11.    Cora B. Excelente-Toledo and Nicholas R. Jennings, (2003), Learning When and How to Coordinate Web Intelligence and Agent System, ,203-218, .
12.    H. Denis, (2007), Managing collaborative learning processes in e-learning applications, in: 29th International Conference on Information Technology Interfaces, pages. 345-350.
13.    Jochems, W., Van Merrienboer, J, Koper, R., & Van Merrienboer, J. G. (2003), Integrated E-Learning: Pedagogy, technology, and organization (Open and Flexible Learning Series), NY: Kogan
14.    Nielsen, J., (1993), Usability Engineering. San Diego,CA: Morgan Kaufmann.
15.    Xiaoyuan Su  and Taghi M.  Khoshgoftaar (2009),, A survey of collaborative filtering techniques in: Advances in Artificial Intelligence.
16.    UM 97 Reader's Guide. User Modeling: Proceedings of the Sixth International Conference, UM97. On-line proceedings, 1997. Available from the World Wide Web: http://um.org/
17.    Andresen,K., (2006), Design and Use Patterns of adaptability in Enterprise Systems Gito, Berlin.
18.    Mirko Morandini , Loris Penserini and Anna PeriniTowards(2008), goal-oriented development of self-adaptive systems in : international Conference on Software Engineering Proceedings of the 2008 international workshop on Software engineering for adaptive and self-managing systems Leipzig, Germany pages 9,16
19.    Angelique Dimitracopoulou, (2005), Designing collaborative learning systems: current trends & future research agenda Computer Support for Collaborative Learning  Proceedings of the 2005 conference on Computer support for collaborative learning: learning the next 10 year Taipei, Taiwan Pages: 115 - 124  
20.    A System for Building Intelligent Agents that Learn to Retrieve and Extract Information in User Modeling and User-Adapted Interaction ,  February -May 2003 Pages: 35 - 88 
21.    Yi-Hsing Chang Tsung-Yi Lu and Rong-Jyue Fang, An adaptive e-learning system based on intelligent agents Proceedings of the 6th Conference on WSEAS
22.    I. Ivan and C. Ciurea. (2008, December 10), Va-lidations of metrics for collaborative systems. Infor-matica Economică Journal in :International Conference on Applied Computer Science.
23.    Niţchi, R. Niţchi and A. Mihăilă. (2007, Sep-tember 15), Some Remarks on Collaborative Systems Framework. Informatica Economică Journal.
 

Tagged as:

No tags for this article