Mara Abel

Enhancing digital libraries with TechLens

The number of research papers available is growing at a staggering rate. Researchers need tools to help them find the papers they should read among all the papers published each year. In this paper, we present and experiment with hybrid recommender algorithms that combine collaborative filtering and content-based filtering to recommend research papers to users. Our hybrid algorithms combine the strengths of each filtering approach to address their individual weaknesses. We evaluated our algorithms through offline experiments on a database of 102,000 research papers, and through an online experiment with 110 users. For both experiments we used a dataset created from the CiteSeer repository of computer science research papers. We developed separate English and Portuguese versions of the interface and specifically recruited American and Brazilian users to test for cross-cultural effects. Our results show that users value paper recommendations, that the hybrid algorithms can be successfully combined, that different algorithms are more suitable for recommending different kinds of papers, and that users with different levels of experience perceive recommendations differently. These results can be applied to develop recommender systems for other types of digital libraries.

PetroGrapher: managing petrographic data and knowledge using an intelligent database application

This paper describes the PetroGrapher system, an intelligent database application to support petrographic analysis, interpretation of oil reservoir rocks, and management of relevant data using resources from both knowledge-based system technology and database technology. In this project, the visual tacit knowledge applied in petrographic analysis was rendered explicit through the collection of cases (rock descriptions), which were then used in the development of a domain ontology organized in a partonomy. Expert-level basic features, which we call ‘visual chunks’, were identified. The cases were further compared against the ontology to elucidate the relations between features in descriptions of rocks, visual chunks and expert interpretations. The domain knowledge was represented through a set of frames and knowledge graphs. The knowledge graphs are applied to recognize the visual chunks in the user data and retrieved the related interpretation. The system was developed as a structure tightly coupled with a relational database system, which acts as a repository for the knowledge base and the user data, and an object-oriented component, which preserves the semantics of data and develops inferences. The system was validated by three groups of users with different levels of expertise. q 2003 Elsevier Ltd. All rights reserved.

Ontological primitives for visual knowledge

In the last few years, we have analyzed the best alternatives for acquiring and processing visual knowledge with the goal of supporting problem solving. We call visual knowledge the set of mental models that support the process of reasoning over information that comes from the spatial arrangement and visual aspects of entities. Also, visual knowledge is implicit, meaning that it is difficult to be explicitly represented solely with propositional constructs. In this paper, we describe a representational approach that helps geologists in capturing and applying this kind of knowledge, in order to support software development applied to interpretation tasks in Petroleum Geology applications. Our approach combines propositional constructs with visual pictorial constructs in order to model visual knowledge of geologists. These constructs are proposed in a strong formal model, founded by Formal Ontology concepts. Based on these constructs, we develop a full ontology for stratigraphic description of sedimentary facies. The Formal Ontology background and the approach are detailed and evaluated through the paper.

Representing part–whole relations in conceptual spaces

In this paper, we propose a cognitive semantic approach to represent part–whole relations. We base our proposal on the theory of conceptual spaces, focusing on prototypical structures in part–whole relations. Prototypical structures are not accounted for in traditional mereological formalisms. In our account, parts and wholes are represented in distinct conceptual spaces; parts are joined to form wholes in a structure space. The structure space allows systematic similarity judgments between wholes, taking into consideration shared parts and their configurations. A point in the structure space denotes a particular part structure; regions in the space represent different general types of part structures. We argue that the structural space can represent prototype effects: structural types are formed around typical arrangements of parts. We also show how structure space captures the variations in part structure of a given concept across different domains. In addition, we discuss how some taxonomies of part–whole relations can be understood within our framework.

Visual Interpretation of Events in Petroleum Geology

In visual domains, the tasks are accomplished through intensive use of visual knowledge. In this paper, we are interested in the visual interpretation task, which is prevalent in many visual domains. We investigate the role played by foundational ontologies in reasoning processes that deals with visual information, as those that are performed in visual interpretation tasks. We propose an approach for visual interpretation that combines ontologically well founded domain ontologies, a reasoning model, and a cognitively well founded meta-model for representation of inferential knowledge. Our approach was effectively applied in the task of visual interpretation of depositional processes, within the Sedimentary Stratigraphy domain.

Defining positioning in a core ontology for robotics

Unambiguous definition of spatial position and orientation has crucial importance for robotics. In this paper we propose an ontology about positioning. It is part of a more extensive core ontology being developed by the IEEE RAS Working Group on ontologies for robotics and automation. The core ontology should provide a common ground for further ontology development in the field. We give a brief overview of concepts in the core ontology and then describe an integrated approach for representing quantitative and qualitative position information.

PersonalTour: A Recommender System for Travel Packages

This paper describes the Personal Tour recommender system that helps customers to find best travel packages according to their preferences. Personal Tour is based on the paradigm of the Distributed Artificial Intelligence and a customer recommendation request is divided into partial recommendations that are handled by different agents. Experiments were run with real customers and the results are presented.

Ontology for Imagistic Domains: Combining Textual and Pictorial Primitives

This paper proposes a knowledge model for representing concepts that requires pictorial as well as conceptual representation to fully capture the ontological meaning. The model was built from the proposition of pictorial primitives to be associated to the original conceptual primitives. The formalized pictorial content is then used to provide an organization to the domain, based on the visual characteristics of the objects as humans are used to do. The combination of both primitives allows the definition of domain ontologies to support visual interpretation activities. The approach was applied to build the Stratigraphy ontology for the definition of Sedimentary Facies and Structures.

A Density-Based Approach for Instance Selection

Instance selection is an important preprocessing step that can be applied in many machine learning tasks. Due to the increasing of the size of the datasets, techniques for instance selection have been applied for reducing the data to a manageable volume, leading to a reduction of the computational resources that are necessary for performing the learning process. Besides that, algorithms of instance selection can also be applied for removing useless, erroneous or noisy instances, before applying learning algorithms. This step can improve the accuracy in classification problems. In the last years, several approaches for instance selection have been proposed. However, most of them have long runtimes and, due to this, they cannot be used for dealing with large datasets. In this paper, we propose a simple and effective density-based approach for instance selection. Our approach, called LDIS (local density-based instance selection), evaluates the instances of each class separately and keeps only the densest instances in a given (arbitrary) neighborhood. This ensures a reasonably low time complexity. Our approach was evaluated on 15 well-known data sets and its performance was compared with the performance of 5 state-of-the-art algorithms, considering three measures: accuracy, reduction and effectiveness. For evaluating the accuracy achieved using the datasets produced by the algorithms, we applied the KNN algorithm. The results show that LDIS achieves a performance (in terms of balance of accuracy and reduction) that is better or comparable to the performances of the other algorithms considered in the evaluation.

Visual interpretation of events in petroleum exploration

We propose an ontology-based approach for visual interpretation tasks.Our approach applies ontological meta-properties of a foundational ontology.We propose a cognition-inspired representation structure for inferential knowledge.We apply our approach for interpreting depositional processes, in Petroleum Geology. In visual domains, such as Medicine, Meteorology and Geology, the tasks are accomplished through intensive use of visual knowledge. In this work, we focus in an essential kind of task that is performed in many visual domains: the visual interpretation task. We call visual interpretation the expert reasoning process that starts with the perception of visual features of domain objects and results in the understanding of the scene. We propose a top-down approach for solving visual interpretation tasks, which is based on symbolic pattern matching supported by well-founded ontologies. Our approach combines well-founded domain ontologies; a meta-model for representing inferential knowledge, which is based on the notion of perceptual chunks; and a reasoning model for visual interpretation. The proposed model was applied for developing an expert system for automating the task of visual interpretation of depositional processes, which a crucial task for petroleum exploration.