Henrik Legind Larsen

Generalized conjunction/disjunction

The generalized conjunction/disjunction function (GCD) is a continuous logic function of two or more variables that integrates conjunctive and disjunctive properties in a single function. It is used as a mathematical model of simultaneity and replaceability of inputs. Special cases of this function include the full (pure) conjunction, the partial conjunction, the arithmetic mean, the partial disjunction, and the full (pure) disjunction. GCD enables a continuous transition from the full conjunction to the full disjunction, using a parameter that specifies a desired level of conjunction (andness) or disjunction (orness). In this paper, we investigate and compare various versions of GCD and other mathematical models of simultaneity and replaceability that are applicable in the areas of system evaluation, and information retrieval.

A fuzzy genetic algorithm approach to an adaptive information retrieval agent

We present an approach to a Genetic Information Retrieval Agent Filter (GIRAF) for documents from the Internet using a genetic algorithm (GA) with fuzzy set genes to learn the users information needs. The population of chromosomes with fixed length represents such users preferences. Each chromosome is associated with a fitness that may be considered the systems belief in the hypothesis that the chromosome, as a query, represents the users information needs. In a chromosome, every gene characterizes documents by a keyword and an associated occurrence frequency, represented by a certain type of a fuzzy subset of the set of positive integers. Based on the users evaluation of the documents retrieved by the chromosome, compared to the scores computed by the system, the fitness of the chromosomes is adjusted. A prototype of GIRAF has been developed and tested. The results of the test are discussed, and some directions for further works are pointed out.

The use of fuzzy relational thesauri for classificatory problem solving in information retrieval and expert systems

The problem-solving strategy applied in knowledge-based systems may often be characterized as classification. Central to classification is computation of the degree to which an object is an instance of a given class (concept, category). Two kinds of problems, namely object-querying and class-querying, as exemplified by, respectively, information retrieval systems and expert systems, are distinguished. In the first kind, the problem is to identify the objects (e.g. documents) to which a given concept (the query) applies. In the second kind, the problem is to identify the concepts (categories) that apply to a given object (the observation). A fuzzy-set-based scheme for construction of efficient problem solving systems of the two kinds is developed. The problem of vocabulary mismatch in information retrieval is considered, and the scheme is proposed as a solution to this problem. The knowledge base applies a term-centered representation form called a fuzzy relational thesaurus. To avoid recomputation of deductive information in problem-solving tasks, the deductive closure of the knowledge base is derived at the outset. This closure is computed in O(n/sup 3/) time. >

Detecting Hidden Hierarchy in Terrorist Networks: Some Case Studies

This paper provides a novel algorithm to automatically detect the hidden hierarchy in terrorist networks. The algorithm is based on centrality measures used in social network analysis literature. The advantage of such automatic methods is to detect key players in terrorist networks. We illustrate the algorithm over some case studies of terrorist events that have occurred in the past. The results show great promise in detecting high value individuals.

On discovering potential inconsistencies in validating uncertain knowledge bases by reflecting on the input

The problem of finding potential inconsistencies during the process of validating knowledge based systems is discussed. A novel methodology, called reflecting on the inputs, is introduced. This effectively requires the knowledge base to stipulate the possible values for the inputs to the system. It is shown that the existence of potential inconsistencies manifest themselves by putting restrictions on the allowable input values. >

Notice of Violation of IEEE Publication Principles Practical approaches for analysis, visualization and destabilizing terrorist networks

Traditionally most of the literature in social network analysis (SNA) has focused on networks of individuals. Although SNA is not conventionally considered as a data mining technique, it is especially suitable for mining a large volume of association data to discover hidden structural patterns in terrorist networks. After September 11 attacks, SNA has increasingly been used to study terrorist networks. As these covert networks share some features with conventional networks, they are harder to identify because they mask their transactions. The most complicating factor is that terrorist networks are often embedded in a much larger population (i.e., adversaries have links with both covert and innocent individuals). Hence, it is desirable to have tools to correctly classify individuals in covert networks so that the resources for isolating them will be used more efficiently. This paper uses centrality measures from complex networks to discuss how to destabilize adversary networks. We propose newly introduced algorithms for constructing hierarchy of the covert networks, so that investigators can view the structure of the ad hoc networks/atypical organizations, in order to destabilize the adversaries. The algorithms are also demonstrated by using publicly available dataset. Moreover we also demonstrate techniques for filtering graphs (networks)/detecting particular cells in adversary networks using a fictitious dataset.

Importance weighted OWA aggregation of multicriteria queries

The family of ordered weighted averaging (OWA) operators, as introduced by R.R. Yager (1988), appears to be very useful in flexible query answering, including information retrieval, where the information needs are often modeled by an aggregation (of the criteria in the query) between /spl and/ (pure AND) and V (pure OR). In this paper, we discuss and set up a general set of requirements for importance-weighted aggregation, and show that an importance weighting scheme suggested by Yager (1977) for OWA operators satisfies the requirements. The weighted arithmetic mean is shown to be order-equivalent to the special case of importance-weighted OWA operators where the importance-weighted satisfaction of the criteria are weighted evenly in the OWA aggregation.

Practical Algorithms for Destabilizing Terrorist Networks

This paper uses centrality measures from complex networks to discuss how to destabilize terrorist networks. We propose newly introduced algorithms for constructing hierarchy of covert networks, so that investigators can view the structure of terrorist networks / non-hierarchical organizations, in order to destabilize the adversaries. Based upon the degree centrality, eigenvector centrality, and dependence centrality measures, a method is proposed to construct the hierarchical structure of complex networks. It is tested on the September 11, 2001 terrorist network constructed by Valdis Krebs. In addition we also propose two new centrality measures i.e., position role index (which discovers various positions in the network, for example, leaders / gatekeepers and followers) and dependence centrality (which determines who is depending on whom in a network). The dependence centrality has a number of advantages including that this measure can assist law enforcement agencies in capturing / eradicating of node (terrorist) which may disrupt the maximum of the network.

Efficient andness-directed importance weighted averaging operators

Importance weighted averaging is a central information processing task in multicriteria decision problems of many kinds, such that selection, classification, object recognition, query answering, and information retrieval. These problems are characterized by a query, i.e., a set of importance weighted criteria, and a set of options queried. While each criterion determines a ranking of the options, the task of the averaging operator is essentially to aggregate these rankings into an overall ranking under the consideration of the criterion importance. We present a class of such operators, based on the power means, namely the Andness-directed Importance Weighted Averaging (AIWA) operators. The operators are equipped with an approximate andness measure allowing an easy, direct control of the andness in the unit interval. The aggregation behavior of the operators appears to be similar to that of importance weighted maximum entropy OWA operators. However, AIWA operators aggregates n arguments (criterion satisfaction values) in O(n) time as opposed to O(n log n) time for OWA operators. An interesting property provided by AIWA operators is decomposability, allowing us to consider new or improved criteria without recomputing with all arguments. Overall, the AIWA operators appear to be effective as andness controlled, importance weighted averaging operators, as well as easy to apply and computationally efficient.

A framework for fuzzy recognition technology

Presents a scheme for object recognition by classificatory problem solving in the framework of fuzzy sets and possibility theory. The scheme has a particular focus on handling the imperfection problems that are common in application domains where the objects to be recognized (detected and identified) represent undesirable situations, referred to as crises. Crises develop over time, and observations typically increase in number and precision as the crisis develops. Early detection and precise recognition of crises is desired, since it increases the possibility of an effective treatment. The crisis recognition problem is central in several areas of decision support, such as medical diagnosis, financial decision making and early warning systems. The problem is characterized by vague knowledge and observations suffering from several kinds of imperfections, such as missing information, imprecision, uncertainty, unreliability of the source, and mutual (possibly conflicting or reinforcing) observations of the same phenomena. The problem of handling possibly imperfect observations from multiple sources includes the problems of information fusion and multiple-sensor data fusion. The different kinds of imperfection are handled in the framework of fuzzy sets and possibility theory.