Aurelio López-López

The segmented and annotated IAPR TC-12 benchmark

Automatic image annotation (AIA), a highly popular topic in the field of information retrieval research, has experienced significant progress within the last decade. Yet, the lack of a standardized evaluation platform tailored to the needs of AIA, has hindered effective evaluation of its methods, especially for region-based AIA. Therefore in this paper, we introduce the segmented and annotated IAPR TC-12 benchmark; an extended resource for the evaluation of AIA methods as well as the analysis of their impact on multimedia information retrieval. We describe the methodology adopted for the manual segmentation and annotation of images, and present statistics for the extended collection. The extended collection is publicly available and can be used to evaluate a variety of tasks in addition to image annotation. We also propose a soft measure for the evaluation of annotation performance and identify future research areas in which this extended test collection is likely to make a contribution.

Information Retrieval with Conceptual Graph Matching

The use of conceptual graphs for the representaion of text contents in information retrievel is discussed. A method for measuring the similarity between two texts represented as conceptual graphs is presented. The method is based on well-known strategies of text comparison, such as Dice coefficient, with new elements introduced due to the bipartite nature of the conceptual graphs. Examples of the representation and comparison of the phrases are given. The structure of an information retrieval system using two-level document representation, traditional keywords and conceptual graphs, is presented.

Flexible Comparison of Conceptual Graphs

Conceptual graphs allow for powerful and computationally affordable representation of the semantic contents of natural language texts. We propose a method of comparison (approximate matching) of conceptual graphs. The method takes into account synonymy and subtype/supertype relationships between the concepts and relations used in the conceptual graphs, thus allowing for greater flexibility of approximate matching. The method also allows the user to choose the desirable aspect of similarity in the cases when the two graphs can be generalized in different ways. The algorithm and examples of its application are presented. The results are potentially useful in a range of tasks requiring approximate semantic or another structural matching - among them, information retrieval and text mining.

Comparison of Conceptual Graphs

In intelligent knowledge-based systems, the task of approximate matching of knowledge elements has crucial importance. We present the algorithm of comparison of knowledge elements represented with conceptual graphs. The method is based on well-known strategies of text comparison, such as Dice coefficient, with new elements introduced due to the bipartite nature of the conceptual graphs. Examples of comparison of two pieces of knowledge are presented. The method can be used in both semantic processing in natural language interfaces and for reasoning with approximate associations.

A language independent method for question classification

Previous works on question classification are based on complex natural language processing techniques: named entity extractors, parsers, chunkers, etc. While these approaches have proven to be effective they have the disadvantage of being targeted to a particular language. We present here a simple approach that exploits lexical features and the Internet to train a classifier, namely a Support Vector Machine. The main feature of this method is that it can be applied to different languages without requiring major modifications. Experimental results of this method on English, Italian and Spanish show that this approach can be a practical tool for question answering systems, reaching a classification accuracy as high as 88.92%.

Text Mining at Detail Level Using Conceptual Graphs

Text mining is defined as knowledge discovery in large text collections. It detects interesting patterns such as clusters, associations, deviations, similarities, and differences in sets of texts. Current text mining methods use simplistic representations of text contents, such as keyword vectors, which imply serious limitations on the kind and meaningfulness of possible discoveries. We show how to do some typical mining tasks using conceptual graphs as formal but meaningful representation of texts. Our methods involve qualitative and quantitative comparison of conceptual graphs, conceptual clustering, building a conceptual hierarchy, and application of data mining techniques to this hierarchy in order to detect interesting associations and deviations. Our experiments show that, despite widespread misbelief, detailed meaningful mining with conceptual graphs is computationally affordable.

Toward a Document Model for Question Answering Systems

The problem of acquiring valuable information from the large amounts available today in electronic media requires automated mechanisms more natural and efficient than those already existing. The trend in the evolution of information retrieval systems goes toward systems capable of answering specific questions formulated by the user in her/his language. The expected answers from such systems are short and accurate sentences, instead of large document lists. On the other hand, the state of the art of these systems is focused -mainly- in the resolution of factual questions, whose answers are named entities (dates, quantities, proper nouns, etc). This paper proposes a model to represent source documents that are then used by question answering systems. The model is based on a representation of a document as a set of named entities (NEs) and their local lexical context. These NEs are extracted and classified automatically by an off-line process. The entities are then taken as instance concepts in an upper ontology and stored as a set of DAML+OIL resources which could be used later by question answering engines. The paper presents a case of study with a news collection in Spanish and some preliminary results.

Detecting Deviations in Text Collections: An Approach Using Conceptual Graphs

Deviation detection is an important problem of both data and text mining. In this paper we consider the detection of deviations in a set of texts represented as conceptual graphs. In contrast with statistical and distance-based approaches, the method we propose is based on the concept of generalization and regularity. Among its main characteristics are the detection of rare patterns (that attempt to give a generalized description of rare texts) and the ability to discover local deviations (deviations at different contexts and generalization levels). The method is illustrated with the analysis of a set of computer science papers.

Representing Context Information for Document Retrieval

The bag of words representation (BoW), which is widely used in information retrieval (IR), represents documents and queries as word lists that do not express anything about context information. When we look for information, we find that not everything is explicitly stated in a document, so context information is needed to understand its content. This paper proposes the use of bag of concepts (BoC) and Holographic reduced representation (HRR) in IR. These representations go beyond BoW by incorporating context information to document representations. Both HRR and BoC are produced using a vector space methodology known as Random Indexing, and allow expressing additional knowledge from different sources. Our experiments have shown the feasibility of the representations and improved the mean average precision by up to 7% when they are compared with the traditional vector space model.

Combining Text Vector Representations for Information Retrieval

This paper suggests a novel representation for documents that is intended to improve precision. This representation is generated by combining two central techniques: Random Indexing; and Holographic Reduced Representations (HRRs). Random indexing uses co-occurrence information among words to generate semantic context vectors that are the sum of randomly generated term identity vectors. HRRs are used to encode textual structure which can directly capture relations between words (e.g., compound terms, subject-verb, and verb-object). By using the random vectors to capture semantic information, and then employing HRRs to capture structural relations extracted from the text, document vectors are generated by summing all such representations in a document. In this paper, we show that these representations can be successfully used in information retrieval, can effectively incorporate relations, and can reduce the dimensionality of the traditional vector space model (VSM). The results of our experiments show that, when a representation that uses random index vectors is combined with different contexts, such as document occurrence representation (DOR), term co-occurrence representation (TCOR) and HRRs, the VSM representation is outperformed when employed in information retrieval tasks.