Douglas W. Oard. The state of the art in text filtering. User Modeling and User-Adapted Interaction, 7(3):141-178, 1997.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....the following must hold: P p or P p # . Nor did we fully implement constraint C1. In this case each step of a proof must be checked by testing whether the negation of the atom is provable with no new steps or with steps that cost less than the proof of the original atom. As suggested in [2], in the case of weighted abduction one should settle for incomplete consistency checking and focus on detecting the inconsistencies that are most likely to arise in the application domain. Instead of implementing constraint C1 above, we prevent the application of abductive rules that would ....

Douglas Appelt and Martha Pollack. Weighted abduction for plan ascription. User Modeling and UserAdapted Interaction, 2(1 -- 2):1 -- 25, 1992.

....and Wellman, 1995] and [Paek and Horvitz, 2000] use Belief Networks to represent the likelihood of possible plans and goals to be attributed to the user. All of these methods, however, require a complete plan library as well as the assignment of probability distributions over the library. [Appelt and Pollack, 1991] and [Goldman et al. 1999] cast plan recognition as weighted abduction. However, this also requires a plan library and the acquisition of weights for abductive rules. Abduction is also computationally hard, and it is unclear whether such routines would be fast enough for complex domains. ....

Douglas E. Appelt and Martha E. Pollack. Weighted abduction for plan ascription. User Modeling and User-Adapted Interaction, 2:1--25, 1991.

....must hold: T [ p or T [ p . Nor did we fully implement constraint C1. In this case each step of a proof must be checked by testing whether the negation of the literal is provable with no new steps or with steps that cost less than the proof of the original literal. As suggested in [2], in the case of weighted abduction one should settle for incomplete consistency checking and focus on detecting the inconsistencies that are most likely to arise in the application domain. Instead of implementing constraint C1 above, we prevent the application of abductive rules that would ....

Douglas Appelt and Martha Pollack. Weighted abduction for plan ascription. User Modeling and UserAdapted Interaction, 2(1 -- 2):1 -- 25, 1992.

....http: www.science.uva.nl cgmsnoek Section 1 Introduction 1 1 Introduction For browsing, searching, and manipulating video documents, an index describing the video content is required. It forms the crux for applications like digital libraries storing multimedia data, or filtering systems [58] which automatically identify relevant video documents based on a user profile. To cater for these diverse applications, the indexes should be rich and as complete as possible. Until now, construction of an index is mostly carried out by documentalists who manually assign a limited number of ....

D.W. Oard. The state of the art in text filtering. User Modeling and UserAdapted Interaction, 7(3):141--178, 1997.

....indexing, video segmentation, multimodal integration, framework. 1 Introduction For browsing, searching, and manipulating video documents, an index describing the video content is required. It forms the crux for applications like digital libraries storing multimedia data, or ltering systems [58] which automatically identify relevant video documents based on a user pro le. To cater for these diverse applications, the indexes should be rich and as complete as possible. Until now, construction of an index is mostly carried out by documentalists who manually assign a limited number of ....

D.W. Oard. The state of the art in text ltering. User Modeling and User-Adapted Interaction, 7(3):141-178, 1997.

....with the TV Scout system presented in Chapter 5, we will place all our examples into the TV domain. 1.1 BASIC CONCEPTS In the following, we give a brief introduction to information filtering. Additional information about this topic can be found in several surveys, e.g. BC92, LT92, OM96, Oar97] The related field of information retrieval is described in several books, e.g. vRi79, Sal83, Bla90, Ing92, FB92, GF98] A broader picture of information seeking in electronic environments can be found in [Mar95] 1.1.1 Information filtering The term information filtering (IF) Den82] has ....

....In Chapter 6, we conclude with a brief summary of the achievements of this dissertation and an outlook to future work. CHAPTER 2 REQUIREMENTS ANALYSIS AND RELATED WORK The grand challenge for information detection systems is to match rapidly changing information with highly variable interests [Oar97] When filtering information, IF systems make predictions about the relevance of objects with respect to the current user. These predictions are made on the basis of the system s internal model of the user, i.e. the user profile. To maximize the quality of their predictions, IF systems try to ....

D. Oard. The state of the art in text filtering. User Modeling and UserAdapted Interaction, 7(3):141-178, 1997.

....multimedia material. It is becoming dicult to track down the material most relevant to an individual s needs and it is becoming expensive to download a resource just to see if it is relevant [2] Some responses to this problem have been in the form of improved tools for search and retrieval [5], but it has also been shown to be bene cial to provide mechanisms through which users are supported in assessing for themselves the potential relevance of an interesting resource. Techniques such as sentence selection [7] and natural language generation [4] are being used to generate summaries ....

Oard, D.: The State of the Art in Text Filtering. User Modeling and User-Adapted Interaction. 7 (1997) 141-178

....generation of fitter agents. Thereby, the Web pages presented by Amalthaea can converge to the user s taste. Fab[3] is a multi agent system that retrieves and filters Web pages on behalf of the users. Fab s filtering agent employs both content based filtering as well as collaborative filtering. [17] For content based filtering, Fab is based on the vector space model[20] for knowledge representation and filtering. Learning is based on the users relevance feedback and the Rocchio method[18] that updates the weights in a keyword vector, which represent the user s information needs. The ....

D.W. Oard. The state of the art in text filtering. User Modeling and User-Adapted Interaction, 7(3):141--178, 1997.

....plans (a list of user actions) is done through the use of a library of plans and actions, some heuristic rules and the user possible goals. This approach has been used by Litman and Allen ( 7, 6] in order to infer plans behind speech acts in dialogues. A di erent approach was followed by Pollack ([11, 1]) which models plan as mental states and tries to abduct the mental attitudes behind each speach act. In this paper we follow a general approach which will allow us to handle both models. Since we needed non monotonic reasoning, namely default and abductive reasoning, as the basic inference ....

....to be able to read them. A: Even if you set the protections, the system manager can override them. In these examples, the plan ascription is made through the inference of the agents beliefs and intentions. As epistemic operators (describing the agents mental states) we ve de ned the following ([1, 3]) INT(a, agent a intends to do BEL(a, p) agent a beliefs that p is true ACH(a, p) agent a beliefs p will be true as a consequence of its actions EXP(a, p) BEL(a, p) or ACH(a, p) More complex actions can be constructed from the operators TO e BY : TO( p) the plan of performing ....

Douglas E. Appelt and Martha E. Pollack. Weighted abduction for plan ascription. User Modeling and User-Adapted Interaction, 2(1), 1992.

....plans (a list of user actions) is done through the use of a library of plans and actions, some heuristic rules and the user possible goals. This approach has been used by Litman and Allen ( 6, 5] in order to infer plans behind speech acts in dialogues. A di erent approach was followed by Pollack ([10, 1]) which models plan as mental states and tries to abduce the mental attitudes behind each speech act. In this paper we follow a general approach which will allow us to handle both models. This approach uses non monotonic reasoning, namely default Owns a scholarship from JNICT, reference n o ....

.... A: She s already been discharged. Her home number is 555 8321. Accordingly with Pollack s approach, the plan ascription is made through the inference of the agents beliefs and intentions. As epistemic operators (describing the agents mental states) there are de ned the following operators ([1, 2]) INT(a, agent a intends to do BEL(a, p) agent a beliefs that p is true ACH(a, p) agent a beliefs p will be true as a consequence of its actions EXP(a, p) BEL(a, p) or ACH(a, p) More complex actions can be constructed from the operators TO and BY : TO( p) the plan of performing ....

Douglas E. Appelt and Martha E. Pollack. Weighted abduction for plan ascription. User Modeling and User-Adapted Interaction, 2(1), 1992.

....sub actions. The inference of plans is done through the use of a library of plans and actions, some heuristic rules and the user possible goals. This approach has been used by Litman and Allen ( 9, 8] in order to infer plans behind speech acts in dialogues. Another approach has been followed by [2] and uses weighted abductive reasoning taking interpretation as an abductive process ( 7] We also take interpretation as abduction but, in contrast to the weighted abduction approach, we propose a general extended logic programming framework which supports the abduction of intentions behind ....

....logical rules. On the other hand, it is important to point out that we have not proved that the proposed transformations into the abductive event calculus are correct and complete. 5 Epistemic Operators The epistemic operators needed for describing the agents mental state are the following ([2, 3]) int(a, agent a intends to do bel(a, p) agent a believes that p is true ach(a, p) agent a believes p will become true as a consequence of its actions exp(a, p) bel(a, p) or ach(a, p) agent a expects that p is true know(a, p) agent a knows property p knowif(a, p) agent ....

[Article contains additional citation context not shown here]

Douglas E. Appelt and Martha E. Pollack. Weighted abduction for plan ascription. User Modeling and User-Adapted Interaction, 2(1), 1992.

....of plans (a list of user actions) is achieved by using a library of plans and actions, some heuristic rules, and the user possible goals. This approach has been used by Litman and Allen [10, 9] in order to infer plans behind speech acts in dialogues. A di erent approach was followed by Pollack [14, 2] which models plans as mental states and tries to abduce the mental attitudes behind each speech act. We propose a general extended logic programming framework which will allow us to handle both models and to deal with the same kind of dialogues that are dealt with in their approaches. The basic ....

....of(X; Y ) holds at(on hand(X; Y ) E) holds at(free(Y ) E) initiates(E; on(X; Y ) act(E; move to top of(X; Y ) holds at(on hand(X; Y ) E) holds at(free(Y ) E) 5. EPISTEMIC OPERATORS As epistemic operators needed to describe the agents mental state, we have de ned the following [2, 4]: 1. int(a, agent a wants action to be done 2. bel(a, p) agent a believes that p is currently true 3. ach(a, p) agent a believes p will be true as a consequence of the actions of some agent (including its own actions) 4. exp(a, p) bel(a, p) or ach(a, p) agent a expects the uent p to ....

[Article contains additional citation context not shown here]

Douglas E. Appelt and Martha E. Pollack. Weighted abduction for plan ascription. User Modeling and User-Adapted Interaction, 2(1), 1992.

....other formalisms have been proposed for dealing with the uncertainty inherent in plan inference, most notably formal argumentation models 5 and approaches based on probabilistic reasoning. Several researchers have captured plan recognition in a formal model of argumentation[KP89] or abduction[AP92, Wae94]. Appelt and Pollock[AP92] use weighted abduction in which weights are assigned to the premises of each rule. The cost of proving a conclusion C is the sum of the costs of proving the premises in a rule whose consequent is C. The cost associated with a premise depends on whether it is true, proven ....

....for dealing with the uncertainty inherent in plan inference, most notably formal argumentation models 5 and approaches based on probabilistic reasoning. Several researchers have captured plan recognition in a formal model of argumentation[KP89] or abduction[AP92, Wae94] Appelt and Pollock[AP92] use weighted abduction in which weights are assigned to the premises of each rule. The cost of proving a conclusion C is the sum of the costs of proving the premises in a rule whose consequent is C. The cost associated with a premise depends on whether it is true, proven from other rules, or ....

Douglas E. Appelt and Martha E. Pollack. Weighted Abduction for Plan Ascription. User Modeling and User-Adapted Interaction, 2(1-2), 1992.

....feedback when faulty plans are recognized [189] Problems with plan recognition have been the management of uncertainty and the prohibitive size of the plan library required for serious applications. Various default reasoning techniques have been applied to the former (e.g. weighted abduction [8]) but the latter difficulty has hardly been addressed (but for an exception, see the PHI system [20] Plan recognition is usually performed [109] under the strong assumptions that 1) the recognizer agent has complete knowledge of the domain, and that 2) the agent whose plan is being inferred has ....

Douglas E. Appelt and Martha E. Pollack. Weighted abduction for plan ascription. User Modeling and User-Adapted Interaction, 2(1-2):1--25, 1992.

....for improvement. The paper concludes by identifying some promising directions for further work that would be compatible with our approach. 1. Introduction The topic tracking problem exhibits strong similarities to what has been called information filtering in the field of information retrieval [4]. In both cases, the goal is to process information objects arriving in a stream from some source based at least in part on observations of the user s reactions to previously seen objects. Two major variants of the text filtering problem exist, one in which hard decisions must be made to accept or ....

Douglas W. Oard. The state of the art in text filtering. User Modeling and User-Adapted Interaction, 7(3), 1997.

....approach will be needed to produce a competitive system. The paper concludes by identifying two promising directions for further work. 1. Introduction The topic tracking problem exhibits strong similarities to what has been called information filtering in the field of information retrieval [4]. In both cases, the goal is to process information objects arriving in a stream from some source based at least in part on observations of the user s reactions to previously seen objects. Two major variants of the text filtering problem exist, one in which hard decisions must be made to accept or ....

Douglas W. Oard. The state of the art in text filtering. User Modeling and User-Adapted Interaction, 7(3), 1997.

....filtering is now more commonly used when the information in question is arriving over a computer network. The vast majority of information filtering research has been focused on filtering electronic text, but interesting work has been done with music, home videos, and other media as well [Oard97b]. Many of the existing text filtering systems require that the user provide an explicit profile which specifies their information needs. What we call adaptive text filtering systems seek to minimize or eliminate this burden by learning the profile automatically. In many research systems, ....

Douglas W. Oard. The state of the art in text filtering. User Modeling and User-Adapted Interaction, 1997. To appear.

....and the Logos Corporation. 1 We used ltc term weights and participated in category B. 2 Current links to most of the references cited in this paper can be found online at http: www.ee. umd.edu medlab filter 2 Adaptive Multilingual Routing We have surveyed text routing techniques elsewhere [6], so here we describe only the technique which we have chosen to apply. Our approach is based on the ranked output paradigm in which the routing system seeks to rank order newly arrived documents with the most useful documents near the top of the list. We have based our work on a technique ....

Douglas W. Oard. The state of the art in text filtering. User Modeling and User-Adapted Interaction, 1997. To appear.

....explicit about the sort of information being retrieved. I will not attempt to draw a sharp distinction between retrieval and filtering in this survey. Although my own work on adaptive cross language text filtering has led me to make this distinction fairly carefully in other presentations (c.f. [21]) such an approach does little to help understand the fundamental techniques which have been applied or the results that have been obtained in this case. Since it is still common to view filtering (detection of useful documents in dynamic document streams) as a kind of retrieval, I will simply ....

Douglas W. Oard. The state of the art in text filtering. User Modeling and User Adapted Interaction, 1997. To appear.

No context found.

Douglas W. Oard. The state of the art in text filtering. User Modeling and User-Adapted Interaction, 7(3):141-178, 1997.

No context found.

Douglas W. Oard. The state of the art in text filtering. User Modeling and User-Adapted Interaction, 7(3):141-178, 1997.

No context found.

D.W. Oard. The state of the art in text filtering. User Modeling and User-Adapted Interaction, (7):141--178, 1997.

No context found.

Douglas E. Appelt and Martha E. Pollack. Weighted abduction for plan ascription. User Modeling and User-Adapted Interaction, 2:1--25, 1991.

No context found.

Douglas Appelt and Martha Pollack. Weighted abduction for plan ascription. User Modeling and User-Adapted Interaction, 1(4), 1991.

No context found.

Douglas E. Appelt and Martha E. Pollack. Weighted abduction for plan ascription. User Modeling and User-Adapted Interaction, 2:1--25, 1991.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC