The difference between typical instances and atypical instances in a natural categorization process has been introduced by E. Rosh and studied by cognitive psychology and AI. A lot of the knowledge representation systems are expressed in using fuzzy concepts but a degree of membership raises some problem for natural categorizations (especially to classification problems in anthropology, ethnology, archeology, linguistics but also in ontologies), but atypical instances of a concept cannot be apprehended adequately by different degrees from a prototype. Other formal approaches, as paraconsistent logics or non monotonic logics, conceptualize often atypical objects as exceptions. It had yet been developed an alternative way with the logics of determination of the objects (LDO). In this paper, we present the logics of typical and atypical (LTA) in order to give directly a logical approach of typicality / atypicality associated to a concept by a more common way than in LDO, in using only classes and not determination operators. It is introduced a distinction between predicative property and concept defined with its intension and its essence, a part of intension. A typical instance of a concept inherits all properties of intension; a typical instance inherits only properties of essence but it is a full member of the category associated to a concept and not a member with a weak degree of membership. In natural categorization, there are often instances (the exceptions) which do not inherit some properties of the essence; they cannot be considered as atypical instance and belong to the boundary of the category.

Information retrieval applications are essential tools to manage the huge amount of information in the Web. Ontologies have great importance in these applications. The idea here is that several data belonging to a domain of interest are represented and related semantically in the ontology, which can help to navigate, manage and reuse these data. Despite of the growing need of ontology, only few works were interested in Arabic language. Indeed, arabic texts are highly ambiguous, especially when diacritics are absent. Besides, existent works does not cover all the types of semantic relations, which are useful to structure Arabic ontologies. A lot of work has been done on cooccurrence- based techniques, which lead to over-generation. In this paper, we propose a new approach for Arabic semantic relation extraction. We use vocalized texts to reduce ambiguities and propose a new distributional approach for similarity calculus, which is compared to cooccurrence. We discuss our contribution through experimental results and propose some perspectives for future research.

This article compares Defeasible Description Logics (DDL) and Topological Approach to reason on Ontologies with exceptions. DDL is integration between Description Logics and Defeasible Logics to deal with monotonic and non-monotonic parts of the knowledge bases respectively. Topological approach tries to reason on inconsistent knowledge bases using the conventional topological operators e.g., interior, exterior, border and closure. We develop neo-Topology based on topological operators and we make major development and improvements of current Topological approach by properly introducing the ``Thickness Border'' with strong inference rules. We proof the validity of the inference rules using set operations. We demonstrate both approaches with appropriate example. We show the advantages and disadvantages of both approaches.

In informational terms, a module dedicated to process information always has specific inputs and outputs. It describes a particular process constrained by specific rules. A processing chain can be a serial combination and/or a parallel combination of such modules. Thus, in an architecture of language engineering, each processing chain becomes a particular instantiation of all possible paths. A processing chain is built from a choice of modules underlying tasks that an engineer wants to apply to the text. In our paper we will present our theoretical model of logical representation of the processing chains, based on combinatory logic and a formal approach based on categorial grammars and applicative grammar, along with many cases of modules configurations.

In this paper we experiment with supervised machine learning techniques for the task of assigning semantic categories to nouns in Czech. The experiments work with 16 semantic categories based on available manually annotated data. The paper compares two possible approaches - one based on the contextual information, the other based upon morphological properties - we are trying to automatically extract final segments of lemmas which might carry semantic information. The central problem of this research is finding the features for machine learning that produce better results for relatively small training data size.

Introduction to the papers presented at the 2013 FLAIRS Conference Special Track on Artificial Intelligence, Cognitive Semantic, Computational Linguistics, and Logic.

Currently, signature based Intrusion Detection Systems (IDS) approaches are inadequate to address threats posed to networked systems by zero-day exploits. Statistical machine learning techniques offer a great opportunity to mitigate these threats. However, at this point, statistical based IDS systems are not mature enough to be implemented in realtime systems and the techniques to be used are not sufficiently understood. This study focuses on a recently expanded corpus for IDS analysis. Feature analysis and Support Vector Machines classification are performed to obtain a better understanding of the corpus and to establish a baseline set of results which can be used by other studies for comparison. Results of the classification and feature analysis are discussed.

Recognizing avatar faces is a challenge and very important issue for terrorism and security experts. Recently some avatar face recognition techniques are proposed but they are still limited. In this paper, we propose a novel face recognition technique based on discrete wavelet transform and Adapted Median Binary Pattern (AMBP) operator to recognize avatar faces from different virtual worlds. The original LBP operator mainly thresholds pixels in a specific predetermined window based on the central pixel’s value of that window. As a result the LBP operator becomes more sensitive to noise especially in near-uniform or flat area regions of an image. One way to reduce the effect of noise is to update the threshold automatically based on all pixels in the neighborhood using some simple statistical operations. Experiments conducted on two virtual world avatar face image datasets show that our technique performs better than original LBP, adapted LBP, Median Binary Pattern (MBP) and wavelet statistical adapted LBP in terms of accuracy.

We present a method for detecting when a user’s remote ac- cess account has been compromised in such a way that an attacker model can be learned during operations. A Naive Bayes model is built for each user that stores the likelihood for each remote session based on a variety of features avail- able in the access logs. During operation, we leverage Ex- pectation Maximization on new data to update both the user and attacker models, based on the likelihood of the observed session, and perform a model comparison to test for compro- mise. The system scales linearly with the number of users in computation and memory. We present experimental results on a medium-sized enterprise network of over two thousand users, performing “masquerade detection” in which the activ- ity of one user is discovered within another user’s logs.

Algorithms for classifying pre-tagged person entities in tweets into one of 8 profession categories are presented. A classifier using a semi-supervised learning algorithm that takes into consideration the local context surrounding the entity in the tweet, hash tag information, and topic signature scores is described. A method that uses data from the Web to dynamically create a reference file called a person dictionary, which contains person/profession relationships, is described, as is an algorithm to use the dictionary to assign a person into one of the 8 profession categories. Results show that classifications made with the automated person dictionary compare favorably to classifications made using a manually compiled dictionary. Results also show that classifications made using either the dictionary or the classifier are moderately successful and that a hybrid method using both offers significant improvement.