Just-in-time information retrieval agents

نویسندگان

  • Bradley J. Rhodes
  • Pattie Maes
چکیده

A just-in-time information retrieval agent (JITIR agent) is software that proactively retrieves and presents information based on a person's local context in an easily accessible yet nonintrusive manner. This paper describes three implemented JITIR agents: the Remembrance Agent, Margin Notes, and Jimminy. Theory and design lessons learned from these implementations are presented, drawing from behavioral psychology, information retrieval, and interface design. They are followed by evaluations and experimental results. The key lesson is that users of JITIR agents are not merely more efficient at retrieving information, but actually retrieve and use more information than they would with traditional search engines. I n this paper, just-in-time information retrieval agents (JITIR agents) are introduced. JITIR agents (pronounced " jitter agents ") are a class of software agents that proactively present information based on a person's context in an easily accessible and non-intrusive manner. They continuously watch a per-son's environment and present information that may be useful without requiring any action on the part of the user. The environment that an agent looks at is usually computational: e-mail, a Web page that a person is reading, or a document that he or she is writing. However, it can also be a person's physical environment as sensed by a hand-held or wearable computer. The information a JITIR agent provides can come from any number of preindexed databases of documents, e.g., e-mail archives, notes files, or documents from commercial databases such as the INSPEC collection of the Institute of Electrical and Electronics Engineers (IEEE) technical paper abstracts. A key feature of a JITIR agent is that it can provide information from an unordered set of documents ; no hand-coding of documents or special domain dependent techniques are required. This feature makes it easy to adapt systems to new domains and information sources. The word " agent " has many different definitions. JITIR agents are " software agents " in that they are long-lived, watch an environment, and can take action based on that environment without direct user intervention. They should not be confused with " embodied conversational agents " or synthetic characters , which are graphical interactive characters that present a personality and interact with a user in an animistic way. They should also not be confused with distributed agent architectures or agent-oriented programming models, which are both architectures for software design rather than a class of applications. Unless specified, all references to the word …

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Reducing Retrieval Time in Automated Storage and Retrieval System with a Gravitational Conveyor Based on Multi-Agent Systems

The main objective of this study is to reduce the retrieval time of a list of products by choosing the best combination of storage and retrieval rules at any time. This is why we start by implementing some storage rules in an Automated Storage/Retrieval System (Automated Storage and Retrieval System: AS/RS) fitted with a gravity conveyor while some of these rules are dedicated to storage and ot...

متن کامل

بررسی تأثیرات ریشه‌یابی در بازیابی اطلاعات در زبان فارسی

Using the language-specific behavior in information retrieval systems can improve the quality of the retrieved results significantly. Part of the word that remains after removing its affixes is called stem. Stemming process can be used for improving the relevancy of the results in information retrieval system. Different morphological variants of words (plural, past tense…) will be mapped into t...

متن کامل

Using Multiagent Profiling for Distributed Information Retrieval

In this paper, we propose a multiagent information retrieval strategy in which, each agent learns from its experience through its interactions with other agents their capabilities and qualifications. Based on a distributed ontology learning framework, our methodology allows an agent to profile other agents in a dynamic translation table and a neighborhood profile, which together help determine ...

متن کامل

Improved Skips for Faster Postings List Intersection

Information retrieval can be achieved through computerized processes by generating a list of relevant responses to a query. The document processor, matching function and query analyzer are the main components of an information retrieval system. Document retrieval system is fundamentally based on: Boolean, vector-space, probabilistic, and language models. In this paper, a new methodology for mat...

متن کامل

Improved Skips for Faster Postings List Intersection

Information retrieval can be achieved through computerized processes by generating a list of relevant responses to a query. The document processor, matching function and query analyzer are the main components of an information retrieval system. Document retrieval system is fundamentally based on: Boolean, vector-space, probabilistic, and language models. In this paper, a new methodology for mat...

متن کامل

Experience-Based Resource Description and Selection in Multiagent Information Retrieval

In this paper, we propose an agent-centric approach to resource description and selection in a multiagent information retrieval (IR). In the multiagent system, each agent learns from its experience through its interactions with other agents their capabilities and qualifications. Based on a distributed ontology learning framework, our methodology allows an agent to profile other agents in a dyna...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:
  • IBM Systems Journal

دوره 39  شماره 

صفحات  -

تاریخ انتشار 2000