Search

International audience

International audience

One of the issues of e-learning web based application is to understand how the learner interacts with an e-learning application to perform a given task. This study proposes a methodology to analyze learner mouse movement in order to infer the task performed. To do this, a Hidden Markov Model is used for modeling the interaction of the learner with an e-learning application. The obtained results show the ability of our model to analyze the interaction in order to recognize the task performed by the learner., Comment: Fourth International Conference on Information and Communication Technology and Accessibility (ICTA), 2013

The approach described here allows using membership function to represent imprecise and uncertain knowledge by learning in Fuzzy Semantic Networks. This representation has a great practical interest due to the possibility to realize on the one hand, the construction of this membership function from a simple value expressing the degree of interpretation of an Object or a Goal as compared to an other and on the other hand, the adjustment of the membership function during the apprenticeship. We show, how to use these membership functions to represent the interpretation of an Object (respectively of a Goal) user as compared to an system Object (respectively to a Goal). We also show the possibility to make decision for each representation of an user Object compared to a system Object. This decision is taken by determining decision coefficient calculates according to the nucleus of the membership function of the user Object.

The main objective of this paper is to develop a new semantic Network structure, based on the fuzzy sets theory, used in Artificial Intelligent system in order to provide effective on-line assistance to users of new technological systems. This Semantic Networks is used to describe the knowledge of an "ideal" expert while fuzzy sets are used both to describe the approximate and uncertain knowledge of novice users who intervene to match fuzzy labels of a query with categories from an "ideal" expert. The technical system we consider is a word processor software, with Objects such as "Word" and Goals such as "Cut" or "Copy". We suggest to consider the set of the system's Goals as a set of linguistic variables to which corresponds a set of possible linguistic values based on the fuzzy set. We consider, therefore, a set of interpretation's levels for these possible values to which corresponds a set of membership functions. We also propose a method to measure the similarity degree between different fuzzy linguistic variables for the partition of the semantic network in class of similar objects to make easy the diagnosis of the user's fuzzy queries.

Pertinence Feedback is a technique that enables a user to interactively express his information requirement by modifying his original query formulation with further information. This information is provided by explicitly confirming the pertinent of some indicating objects and/or goals extracted by the system. Obviously the user cannot mark objects and/or goals as pertinent until some are extracted, so the first search has to be initiated by a query and the initial query specification has to be good enough to pick out some pertinent objects and/or goals from the Semantic Network. In this paper we present a short survey of fuzzy and Semantic approaches to Knowledge Extraction. The goal of such approaches is to define flexible Knowledge Extraction Systems able to deal with the inherent vagueness and uncertainty of the Extraction process. It has long been recognised that interactivity improves the effectiveness of Knowledge Extraction systems. Novice user's queries are the most natural and interactive medium of communication and recent progress in recognition is making it possible to build systems that interact with the user. However, given the typical novice user's queries submitted to Knowledge Extraction Systems, it is easy to imagine that the effects of goal recognition errors in novice user's queries must be severely destructive on the system's effectiveness. The experimental work reported in this paper shows that the use of possibility theory in classical Knowledge Extraction techniques for novice user's query processing is more robust than the use of the probability theory. Moreover, both possibilistic and probabilistic pertinence feedback can be effectively employed to improve the effectiveness of novice user's query processing.

In this paper we present a synthesis of the work performed on two inference algorithms: the Pearl's belief propagation (BP) algorithm applied to Bayesian networks without loops (i.e. polytree) and the Loopy belief propagation (LBP) algorithm (inspired from the BP) which is applied to networks containing undirected cycles. It is known that the BP algorithm, applied to Bayesian networks with loops, gives incorrect numerical results i.e. incorrect posterior probabilities. Murphy and al. [7] find that the LBP algorithm converges on several networks and when this occurs, LBP gives a good approximation of the exact posterior probabilities. However this algorithm presents an oscillatory behaviour when it is applied to QMR (Quick Medical Reference) network [15]. This phenomenon prevents the LBP algorithm from converging towards a good approximation of posterior probabilities. We believe that the translation of the inference computation problem from the probabilistic framework to the possibilistic framework will allow performance improvement of LBP algorithm. We hope that an adaptation of this algorithm to a possibilistic causal network will show an improvement of the convergence of LBP., The International Conference on Computing & e-Systems - 2007

In this paper we present a short survey of fuzzy and Semantic approaches to Knowledge Extraction. The goal of such approaches is to define flexible Knowledge Extraction Systems able to deal with the inherent vagueness and uncertainty of the Extraction process. In this survey we address if and how some approaches met their goal., Comment: arXiv admin note: text overlap with arXiv:1206.1042, arXiv:1206.0925

In this paper, we show the possibility of using a linear Conditional Random Fields (CRF) for terminology extraction from a specialized text corpus., Comment: This paper has been withdrawn by the author due to the poor readability and the low quality of the English

In this paper we present a short survey of fuzzy and Semantic approaches to Knowledge Extraction. The goal of such approaches is to define flexible Knowledge Extraction Systems able to deal with the inherent vagueness and uncertainty of the Extraction process. It has long been recognised that interactivity improves the effectiveness of Knowledge Extraction systems. Novice user's queries is the most natural and interactive medium of communication and recent progress in recognition is making it possible to build systems that interact with the user. However, given the typical novice user's queries submitted to Knowledge Extraction systems, it is easy to imagine that the effects of goal recognition errors in novice user's queries must be severely destructive on the system's effectiveness. The experimental work reported in this paper shows that the use of classical Knowledge Extraction techniques for novice user's query processing is robust to considerably high levels of goal recognition errors. Moreover, both standard relevance feedback and pseudo relevance feedback can be effectively employed to improve the effectiveness of novice user's query processing., Comment: arXiv admin note: substantial text overlap with arXiv:1206.0925, arXiv:1206.1615

The approach described here allows to use the fuzzy Object Based Representation of imprecise and uncertain knowledge. This representation has a great practical interest due to the possibility to realize reasoning on classification with a fuzzy semantic network based system. For instance, the distinction between necessary, possible and user classes allows to take into account exceptions that may appear on fuzzy knowledge-base and facilitates integration of user's Objects in the base. This approach describes the theoretical aspects of the architecture of the whole experimental A.I. system we built in order to provide effective on-line assistance to users of new technological systems: the understanding of "how it works" and "how to complete tasks" from queries in quite natural languages. In our model, procedural semantic networks are used to describe the knowledge of an "ideal" expert while fuzzy sets are used both to describe the approximative and uncertain knowledge of novice users in fuzzy semantic networks which intervene to match fuzzy labels of a query with categories from our "ideal" expert., Comment: arXiv admin note: text overlap with arXiv:1206.1794

Web services are accessed via query interfaces which hide databases containing thousands of relevant information. User's side, distant database is a black box which accepts query and returns results, there is no way to access database schema which reflect data and query meanings. Hence, web services are very autonomous. Users view this autonomy as a major drawback because they need often to combine query capabilities of many web services at the same time. In this work, we will present a new approach which allows users to benefit of query capabilities of many web services while respecting autonomy of each service. This solution is a new contribution in Information Retrieval research axe and has proven good performances on two standard datasets., Comment: ICITES 2012 - 2nd International Conference on Information Technology and e-Services

This paper presents a method to measure the similarity between different fuzzy concepts in order to optimize Semantic networks. The problem approached is the minimization of the time of research and identification of user's Objects and Goals. Indeed, it concerns to determine to each instant the totality of Objects (respectively Goals) among which one can identify rapidly the most satisfactory for the user's Object and Goal. Alone Objects and most similar Goals to Objects and researched Goals of the viewpoint of attribute values will be processed, what will avoid the analysis of all Objects and system Goals far of needs of the user., Comment: 14th International Conference on Systems Science. Wroclaw, Poland, 2001

Recent work in machine learning for information extraction has focused on two distinct sub-problems: the conventional problem of filling template slots from natural language text, and the problem of wrapper induction, learning simple extraction procedures ("wrappers") for highly structured text such as Web pages. For suitable regular domains, existing wrapper induction algorithms can efficiently learn wrappers that are simple and highly accurate, but the regularity bias of these algorithms makes them unsuitable for most conventional information extraction tasks. This paper describes a new approach for wrapping semistructured Web pages. The wrapper is capable of learning how to extract relevant information from Web resources on the basis of user supplied examples. It is based on inductive learning techniques as well as fuzzy logic rules. Experimental results show that our approach achieves noticeably better precision and recall coefficient performance measures than SoftMealy, which is one of the most recently reported wrappers capable of wrapping semi-structured Web pages with missing attributes, multiple attributes, variant attribute permutations, exceptions, and typos., Comment: International Journal of Computational Science - 2008

In this paper we present a synthesis of work performed on tow information retrieval models: Bayesian network information retrieval model witch encode (in) dependence relation between terms and possibilistic network information retrieval model witch make use of necessity and possibility measures to represent the fuzziness of pertinence measure. It is known that the use of a general Bayesian network methodology as the basis for an IR system is difficult to tackle. The problem mainly appears because of the large number of variables involved and the computational efforts needed to both determine the relationships between variables and perform the inference processes. To resolve these problems, many models have been proposed such as BNR model. Generally, Bayesian network models doesn't consider the fuzziness of natural language in the relevance measure of a document to a given query and possibilistic models doesn't undertake the dependence relations between terms used to index documents. As a first solution we propose a hybridization of these two models in one that will undertake both the relationship between terms and the intrinsic fuzziness of natural language. We believe that the translation of Bayesian network model from the probabilistic framework to possibilistic one will allow a performance improvement of BNRM., The International Conference on Computing & e-Systems - 2008

This paper presents a method of optimization, based on both Bayesian Analysis technical and Galois Lattice of Fuzzy Semantic Network. The technical System we use learns by interpreting an unknown word using the links created between this new word and known words. The main link is provided by the context of the query. When novice's query is confused with an unknown verb (goal) applied to a known noun denoting either an object in the ideal user's Network or an object in the user's Network, the system infer that this new verb corresponds to one of the known goal. With the learning of new words in natural language as the interpretation, which was produced in agreement with the user, the system improves its representation scheme at each experiment with a new user and, in addition, takes advantage of previous discussions with users. The semantic Net of user objects thus obtained by learning is not always optimal because some relationships between couple of user objects can be generalized and others suppressed according to values of forces that characterize them. Indeed, to simplify the obtained Net, we propose to proceed to an Inductive Bayesian Analysis, on the Net obtained from Galois lattice. The objective of this analysis can be seen as an operation of filtering of the obtained descriptive graph., Comment: arXiv admin note: text overlap with arXiv:1206.1794