Results 11  20
of
49
Semantic querying of data guided by Formal Concept Analysis
 In Formal Concept Analysis for Artificial Intelligence Workshop at ECAI
, 2012
"... Abstract. In this paper we present a novel approach to handle querying over a concept lattice of documents and annotations. We focus on the problem of “nonmatching documents”, which are those that, despite being semantically relevant to the user query, do not contain the query’s elements and hence c ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
Abstract. In this paper we present a novel approach to handle querying over a concept lattice of documents and annotations. We focus on the problem of “nonmatching documents”, which are those that, despite being semantically relevant to the user query, do not contain the query’s elements and hence cannot be retrieved by typical string matching approaches. In order to find these documents, we modify the initial user query using the concept lattice as a guide. We achieve this by identifying in the lattice a formal concept that represents the user query and then by finding potentially relevant concepts, identified as such through the proposed notion of cousin concepts. Finally, we use a concept semantic similarity metric to order and present retrieved documents. The main contribution of this paper is the introduction of the notion of cousin concepts of a given formal concept followed by a discussion on how this notion is useful for latticebased information indexing and retrieval. 1
A TwoLevel Learning Hierarchy of Concept Based Keyword Extraction for Tag Recommendation
 ECML PKDD Discovery Challenge
, 2009
"... Abstract. Textual contents associated to resources are considered as sources of candidate tags to improve the performance of tag recommenders in social tagging systems. In this paper, we propose a twolevel learning hierarchy of a concept based keyword extraction method to filter the candidate tags a ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
Abstract. Textual contents associated to resources are considered as sources of candidate tags to improve the performance of tag recommenders in social tagging systems. In this paper, we propose a twolevel learning hierarchy of a concept based keyword extraction method to filter the candidate tags and rank them based on their occurrences in concepts existing in the given resources. Incorporating usercreated tags to extract the hidden conceptdocument relationships distinguishes the twolevel from the onelevel learning version, which extracts concepts directly using terms existing in textual contents. Our experiment shows that a multiconcept approach, which considers more than one concept for each resource, improves the performance of a singleconcept approach, which takes into account just the most relevant concept. Moreover, the experiments also prove that the proposed twolevel learning hierarchy gives better performances than one of the onelevel version.
Knowledge discovery in data using formal concept analysis and random projections
 Int. J. Appl. Math. Comp
"... In this paper our objective is to propose a random projections based formal concept analysis for knowledge discovery in data. We demonstrate the implementation of the proposed method on two real world healthcare datasets. Formal Concept Analysis (FCA) is a mathematical framework that offers a concep ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
In this paper our objective is to propose a random projections based formal concept analysis for knowledge discovery in data. We demonstrate the implementation of the proposed method on two real world healthcare datasets. Formal Concept Analysis (FCA) is a mathematical framework that offers a conceptual knowledge representation through hierarchical conceptual structures called concept lattices. However, during the design of a concept lattice, complexity plays a major role.
Computing intensions of digital library collections
 In Proceedings of the the 5th International Conference on Formal Concept Analysis (ICFCA 2007), LNCS 4390
, 2007
"... Abstract. We model a Digital Library as a formal context in which objects are documents and attributes are terms describing documents contents. A formal concept is very close to the notion of a collection: the concept extent is the extension of the collection; the concept intent consists of a set of ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
Abstract. We model a Digital Library as a formal context in which objects are documents and attributes are terms describing documents contents. A formal concept is very close to the notion of a collection: the concept extent is the extension of the collection; the concept intent consists of a set of terms, the collection intension. The collection intension can be viewed as a simple conjunctive query which evaluates precisely to the extension. However, for certain collections no concept may exist, in which case the concept that best approximates the extension must be used. In so doing, we may end up with a too imprecise concept, in case too many documents denoted by the intension are outside the extension. We then look for a more precise intension by exploring 3 different query languages: conjunctive queries with negation; disjunctions of negationfree conjunctive queries; and disjunctions of conjunctive queries with negation. We show that a precise description can always be found in one of these languages for any set of documents. However, when disjunction is introduced, uniqueness of the solution is lost. In order to deal with this problem, we define a preferential criterion on queries, based on the conciseness of their expression. We then show that minimal queries are hard to find in the last 2 of the three languages above. 1
Similarity Measures in Formal Concept Analysis
"... Formal concept analysis (FCA) has been applied successively in diverse fields such as data mining, conceptual modeling, social networks, software engineering, and the semantic web. One shortcoming of FCA, however, is the large number of concepts that typically arise in dense datasets hindering typic ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
Formal concept analysis (FCA) has been applied successively in diverse fields such as data mining, conceptual modeling, social networks, software engineering, and the semantic web. One shortcoming of FCA, however, is the large number of concepts that typically arise in dense datasets hindering typical tasks such as rule generation and visualization. To overcome this shortcoming, it is important to develop formalisms and methods to segment, categorize and cluster formal concepts. The first step in achieving these aims is to define suitable similarity and dissimilarity measures of formal concepts. In this paper we propose three similarity measures based on existent setbased measures in addition to developing the completely novel zerosinduced measure. Moreover, we formally prove that all the measures proposed are indeed similarity measures and investigate the computational complexity of computing them. Finally, an extensive empirical evaluation on realworld data is presented in which the utility and character of each similarity measure is tested and evaluated. 1
Inducing Classes of Terms from Text
"... c ○ SpringerVerlag Abstract. This paper describes a clustering method for organizing in semantic classes a list of terms. The experiments were made using a POS annotated corpus, the ACL Anthology, which consists of technical articles in the field of Computational Linguistics. The method, mainly bas ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
c ○ SpringerVerlag Abstract. This paper describes a clustering method for organizing in semantic classes a list of terms. The experiments were made using a POS annotated corpus, the ACL Anthology, which consists of technical articles in the field of Computational Linguistics. The method, mainly based on some assumptions of Formal Concept Analysis, consists in building bidimensional clusters of both terms and their lexicosyntactic contexts. Each generated cluster is defined as a semantic class with a set of terms describing the extension of the class and a set of contexts perceived as the intensional attributes (or properties) valid for all the terms in the extension. The clustering process relies on two restrictive operations: abstraction and specification. The result is a concept lattice that describes a domainspecific ontology of terms. 1
What Is the Shape of Your Security Policy? Security as a Classification Problem
 NEW SECURITY PARADIGMS WORKSHOP (NSPW)
, 2009
"... This new paradigm defines security policies on causeeffect relations and models security mechanisms in analogy with pattern recognition classifiers. It augments the arsenal of formal computer security evaluation tools with new techniques. A causality model represents possible causes and effects; th ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
This new paradigm defines security policies on causeeffect relations and models security mechanisms in analogy with pattern recognition classifiers. It augments the arsenal of formal computer security evaluation tools with new techniques. A causality model represents possible causes and effects; the causes include threats and the effects may be undesired. Target security policies derived from functional specifications select permitted causalities. Security mechanisms extract features from causes and effects and enforce mechanismspecific policies, approximating the target policy. Advantages of the classifier paradigm are the ability to generalize from incomplete information and examples, to measure classification error and mechanism performance, and to analyze mechanism ensembles and compositions. The classifier paradigm also offers a conception of problem complexity and suggests paying more attention to the impact of mechanisms rather than to their inner workings.
Service Identification Strategies in LegacytoSOA Migration
"... Abstract—LegacytoSOA migration has been extensively researched in the last decade. Numerous approaches have been proposed. However, some of the issues still remain, such as candidate service identification in legacy code, and tool supported (semi)automated and programming language independent ser ..."
Abstract
 Add to MetaCart
Abstract—LegacytoSOA migration has been extensively researched in the last decade. Numerous approaches have been proposed. However, some of the issues still remain, such as candidate service identification in legacy code, and tool supported (semi)automated and programming language independent service extraction. In this research, such existing issues of legacytoSOA migration approaches are addressed. The research initially proposes a consolidated legacytoSOA migration method, which combines the migration feasibility and supporting technology aspects. The research, then aims at investigating the candidate service identification strategies through architectural reconstruction and source code visualization, detection of design patterns and concept analysis. Finally, the research aims at extracting those identified services using codequery technologies enabling the programming language independent service extraction. The overall result of this research is a set of techniques and toolsets that facilitates the (semi)automated legacytoSOA migration. I.
On quantum statistics in data analysis — Extended abstract —
, 802
"... Originally, quantum probability theory was developed to analyze statistical phenomena in quantum systems, where classical probability theory does not apply, because the lattice of measurable sets is not necessarily distributive. On the other hand, it is well known that the lattices of concepts, that ..."
Abstract
 Add to MetaCart
Originally, quantum probability theory was developed to analyze statistical phenomena in quantum systems, where classical probability theory does not apply, because the lattice of measurable sets is not necessarily distributive. On the other hand, it is well known that the lattices of concepts, that arise in data analysis, are in general also nondistributive, albeit for completely different reasons. In his recent book, van Rijsbergen (2004) argues that many of the logical tools developed for quantum systems are also suitable for applications in information retrieval. I explore the mathematical support for this idea on an abstract vector space model, covering several forms of data analysis (information retrieval, data mining, collaborative filtering, formal concept analysis...), and roughly based on an idea from categorical quantum mechanics (Abramsky & Coecke 2004; Coecke & Pavlovic 2007). It turns out that quantum (i.e., noncommutative) probability distributions arise already in this rudimentary mathematical framework. Moreover, a Belltype inequality is formulated for the standard data similarity measures, interpreted in terms version of the vector space model yields easy counterexamples for such inequalities seems to be an indicator of the presence of entanglement, and of a genuine need for quantum statistics in data analysis.
Formal Concept Analysis for Qualitative Data Analysis over Triple Stores
"... Abstract. Business Intelligence solutions provide different means like OLAP, data mining or case based reasoning to explore data. Standard BI means are usually based on mathematical statistics and provide a quantitative analysis of the data. In this paper, a qualitative approach based on a mathemati ..."
Abstract
 Add to MetaCart
Abstract. Business Intelligence solutions provide different means like OLAP, data mining or case based reasoning to explore data. Standard BI means are usually based on mathematical statistics and provide a quantitative analysis of the data. In this paper, a qualitative approach based on a mathematical theory called ”Formal Concept Analysis ” (FCA) is used instead. FCA allows clustering a given set of objects along attributes acting on the objects, hierarchically ordering those clusters, and finally visualizing the cluster hierarchy in socalled Hassediagrams. The approach in this paper is exemplified on a dataset of documents crawled from the SAP community network, which are persisted in a semantic triple store and evaluated with an existing FCA tool called ”ToscanaJ” which has been modified in order to retrieve its data from a triple store. 1 introduction Business Intelligence (BI) solutions provide different means like OLAP, data mining or case based reasoning to explore data. Standard BI means are usually designed to work with numerical data, thus they provide a quantitative analysis of the data (aka ”number crunching”) based on mathematical statistics. In fact, classical BI examples show ”accounting, finance, or some other calculationheavy subject ” [10]. To some extent, though arguably oversimplified, one can understand BI as acting on lists or tables filled with numbers. Compared to number crunching, Formal Concept Analysis (FCA) [3] provides a complementing approach. The starting point of FCA are crosstables (called ”formal contexts”), where the rows stand for some objects, the columns for some attributes, and the cells (intersections of rows and columns) carry the binary information whether an attribute applies to an object (usually indicated by a cross) or not. Based on this crosstable, the objects are clustered to meaningful sets. These clusters form a hierarchy, which can be visually displayed, e.g. by a socalled Hassediagram. A short introduction into FCA, as needed for this paper, is provided in the next section.