Marcelo Ladeira

A multi-agent intelligent environment for medical knowledge

AMPLIA is a multi-agent intelligent learning environment designed to support training of diagnostic reasoning and modelling of domains with complex and uncertain knowledge. AMPLIA focuses on the medical area. It is a system that deals with uncertainty under the Bayesian network approach, where learner-modelling tasks will consist of creating a Bayesian network for a problem the system will present. The construction of a network involves qualitative and quantitative aspects. The qualitative part concerns the network topology, that is, causal relations among the domain variables. After it is ready, the quantitative part is specified. It is composed of the distribution of conditional probability of the variables represented. A negotiation process (managed by an intelligent MediatorAgent) will treat the differences of topology and probability distribution between the model the learner built and the one built-in in the system. That negotiation process occurs between the agents that represent the expert knowledge domain (DomainAgent) and the agent that represents the learner knowledge (LearnerAgent).

UnBBayes: Modeling Uncertainty for Plausible Reasoning in the Semantic Web

The same assumptions that were essential in the document web are still applied for the Semantic Web (SW). They are radical notions of information sharing, which include [Allemang & Hendler, 2008]: (i) the Anyone can say Anything about Any topic (AAA) slogan; (ii) the open world assumption, i.e. there might exist more information out there that we are not aware of, and (iii) nonunique naming, meaning that different people can assign different names to the same concept. However, the Semantic Web differs from its predecessors in the sense that it intends to provide an environment not only for allowing information sharing but also for making it possible to have the effect of knowledge synergy. Nevertheless, this can lead to a chaotic scenario with disagreements and conflicts. We call an environment characterized by the above assumptions a Radical Information Sharing (RIS) environment. The challenge facing SW architects is therefore to avoid the natural chaos to which RIS environments are prone, and move to a state characterized by information sharing, cooperation and collaboration. According to [Allemang & Hendler, 2008], one solution to this challenge lies in modeling. Modeling is a simplified abstraction of some real world phenomenon, which, amongst other things, allows the organizing of information for the community use. Modeling supports information sharing in three ways: it provides a means for human communication, it provides a way for explaining conclusions, and it provides the managing of different viewpoints. There is an immense variety of modeling approaches. In this chapter we will go over a few of these approaches, showing how they can be used and their main limitations related to achieving the full potential of the Semantic Web. First we will show how to apply Unified Modeling Language (UML) [Rumbaugh et al., 1998] and Entity/Relationship (ER) [Chen, 1976] diagrams for modeling. Then we will present Knowledge Representation and Reasoning (KR&R) [Brachman & Levesque, 2004] and describe how KR&R overcomes some of the limitations of UML and ER. Finally, we present Ontology and the Semantic Web [Berners-Lee, 1999] and discuss how it differs from and moves beyond the previous approaches.

Probabilistic ontology and knowledge fusion for procurement fraud detection in Brazil

To cope with societys demand for transparency and corruption prevention, the Brazilian Office of the Comptroller General (CGU) has carried out a number of actions, including: awareness campaigns aimed at the private sector; campaigns to educate the public; research initiatives; and regular inspections and audits of municipalities and states. Although CGU has collected information from hundreds of different sources - Revenue Agency, Federal Police, and others - the process of fusing all this data has not been efficient enough to meet the needs of CGUs decision makers. Therefore, it is natural to change the focus from data fusion to knowledge fusion. As a consequence, traditional syntactic methods must be augmented with techniques that represent and reason with the semantics of databases. However, commonly used approaches fail to deal with uncertainty, a dominant characteristic in corruption prevention. This paper presents the use of Probabilistic OWL (PR-OWL) to design and test a model that performs information fusion to detect possible frauds in procurements involving Federal money. To design this model, a recently developed tool for creating PR-OWL ontologies was used with support from PR-OWL specialists and careful guidance from a fraud detection specialist from CGU.

A GUI Tool for Plausible Reasoning in the Semantic Web using MEBN

As the work with semantics and services grows more ambitious in the semantic Web community, there is an increasing appreciation on the need for principled approaches for representing and reasoning under uncertainty. Reacting to this trend, the World Wide Web Consortium (W3C) has created the Uncertainty Reasoning for the World Wide Web Incubator Group (URW3-XG) to better define the challenge of reasoning with and representing uncertain information available through the World Wide Web and related WWW technologies. In according to the URW3-XG effort this paper presents the implementation of a graphical user interface for building probabilistic ontologies, an application programming interface for saving and loading these ontologies and a proposal to specify formulas for creating conditional probabilistic tables dynamically. The language used for building probabilistic ontologies is probabilistic OWL (Pr-OWL), an extension for OWL based on multi-entity Bayesian network (MEBN).

The Role of Text Pre-processing in Opinion Mining on a Social Media Language Dataset

This work describes an opinion mining application over a dataset extracted from the web and composed of reviews with several Internet slangs, abbreviations and typo errors. Opinion mining is a study field that tries to identify and classify subjectivity, such as opinions, emotions or sentiments in natural language. In this research, 759.176 Portuguese reviews were extracted from the app store Google Play. Due to the large amount of reviews, large-scale processing techniques were needed, involving powerful frameworks such as Hadoop and Mahout. Based on tests conducted it was concluded that pre-processing has an insignificant role in opinion mining task for the specific domain of reviews of mobile apps. The work also contributed to the creation of a corpus consisting of 759 thousand reviews and a dictionary of slangs and abbreviations commonly used in the Internet.

Deep Learning Anomaly Detection as Support Fraud Investigation in Brazilian Exports and Anti-Money Laundering

Normally exports of goods and products are transactions encouraged by the governments of countries. Typically these incentives are promoted by tax exemptions or lower tax collections. However, exports fraud may occur with objectives not related to tax evasion, for example money laundering. This article presents the results obtained in implementing the unsupervised Deep Learning model to classify Brazilian exporters regarding the possibility of committing fraud in exports. Assuming that the vast majority of exporters have explanatory features of their export volume which interrelate in a standard way, we used the AutoEncoder to detect anomalous situations with regards to the data pattern. The databases used in this work come from exports of goods and products that occurred in Brazil in 2014, provided by the Secretariat of Federal Revenue of Brazil. From attributes that characterize export companies, the model was able to detect anomalies in at least twenty exporters.

Exploring the combination of software visualization and data clustering in the software architecture recovery process

Modernizing a legacy system is a costly process that requires deep understanding of the system architecture and its components. Without an understanding of the software architecture that will be rewritten, the entire process of reengineering can fail. For this reason, semi-automatic and automatic techniques for architecture recovery have been active focuses of research. However, there are still important improvements that need to be addressed on this field of research w.r.t. achieving a more accurate architecture recovery process. In this work, we propose to explore if an approach where visualization and clustering applied together can provide a higher accuracy on the software architecture recovery process. An experimental study was conducted in a industrial environment to empirically evaluate our investigation in four commercial systems. Our results indicated a statistically significant increase in the accuracy of the recovered architectures in all cases.

UMP-ST Plug-in: Documenting, Maintaining and Evolving Probabilistic Ontologies Using UnBBayes Framework

Several approaches have been proposed for dealing with uncertainty in the Semantic Web SW. Although probabilistic ontologies PO is one of the most promising approach to model uncertainty in ontologies, no support has been offered to ontological engineers on how to create this more complex type of ontologies. This task has proven to be extremely difficult and hard, which motivated the creation of the Uncertainty Modeling Process for Semantic Technologies UMP-ST, a process that guides users in modeling POs. This paper presents the UMP-ST plug-in, a tool that implements this process and shows how the plug-in, implemented in UnBBayes Framework, overcomes the main problems on modeling probabilistic ontologies: the complexity in creating; the difficulty in maintaining and evolving; and the lack of a centralized tool for documenting these ontologies. The probabilistic ontology for Procurement Fraud Detection and Prevention in Brazil is used to show how the UMP-ST plug-in overcomes these problems. This probabilistic ontology is a proof-of-concept use case created as part of a research project at the Brazilian Office of the Comptroller General CGU. A short version of this paper was presented on the URSW 2013i¾?[3].

An Evaluation of Data Model for NoSQL Document-Based Databases

NoSQL databases offer flexibility in the data model. The document-based databases may have some data models built with embedded documents, and others made with referenced documents. The challenge lies in choosing the structure of the data. This paper proposes a study to analyze if different data models can have an impact on the performance of database queries. To this end, we created three data models: embedded, referenced, and hybrid. We ran experiments on each data model in a MongoDB cluster, comparing the response time of 3 different queries in each model. Results showed a disparity in performance between the data models. We also evaluated the use of indexes in each data model. Results showed that, depending on the type of query and field searched some types of indexes presented higher performance compared to others. Additionally, we carried out an analysis of the space occupied on the storage disk. This analysis shows that the choice of model also affects disk space for storing data and indexes.

Mining ENADE Data from the Ulbra Network Institution

The National Institute of Educational Research and Studies (INEP) provides ENADE data for Higher Education Institutions (IES) from Brazil. This data is a rich source of support in improving the quality of education offered by these IES, but requires the application of data mining techniques to achieve the standards of the learning process and thus achieve improved academic performance of students in different courses. This paper aims to present the steps of mining the data provided by INEP, which will enable the identification of standards for the IES analyzed, as well as serve as a guide for other IES that wish to follow a similar process.