Celso G. Camilo-Junior

SentiHealth-Cancer: A sentiment analysis tool to help detecting mood of patients in online social networks

BACKGROUND Cancer is a critical disease that affects millions of people and families around the world. In 2012 about 14.1 million new cases of cancer occurred globally. Because of many reasons like the severity of some cases, the side effects of some treatments and death of other patients, cancer patients tend to be affected by serious emotional disorders, like depression, for instance. Thus, monitoring the mood of the patients is an important part of their treatment. Many cancer patients are users of online social networks and many of them take part in cancer virtual communities where they exchange messages commenting about their treatment or giving support to other patients in the community. Most of these communities are of public access and thus are useful sources of information about the mood of patients. Based on that, Sentiment Analysis methods can be useful to automatically detect positive or negative mood of cancer patients by analyzing their messages in these online communities. OBJECTIVE The objective of this work is to present a Sentiment Analysis tool, named SentiHealth-Cancer (SHC-pt), that improves the detection of emotional state of patients in Brazilian online cancer communities, by inspecting their posts written in Portuguese language. The SHC-pt is a sentiment analysis tool which is tailored specifically to detect positive, negative or neutral messages of patients in online communities of cancer patients. We conducted a comparative study of the proposed method with a set of general-purpose sentiment analysis tools adapted to this context. METHODS Different collections of posts were obtained from two cancer communities in Facebook. Additionally, the posts were analyzed by sentiment analysis tools that support the Portuguese language (Semantria and SentiStrength) and by the tool SHC-pt, developed based on the method proposed in this paper called SentiHealth. Moreover, as a second alternative to analyze the texts in Portuguese, the collected texts were automatically translated into English, and submitted to sentiment analysis tools that do not support the Portuguese language (AlchemyAPI and Textalytics) and also to Semantria and SentiStrength, using the English option of these tools. Six experiments were conducted with some variations and different origins of the collected posts. The results were measured using the following metrics: precision, recall, F1-measure and accuracy RESULTS The proposed tool SHC-pt reached the best averages for accuracy and F1-measure (harmonic mean between recall and precision) in the three sentiment classes addressed (positive, negative and neutral) in all experimental settings. Moreover, the worst accuracy value (58%) achieved by SHC-pt in any experiment is 11.53% better than the greatest accuracy (52%) presented by other addressed tools. Finally, the worst average F1 (48.46%) reached by SHC-pt in any experiment is 4.14% better than the greatest average F1 (46.53%) achieved by other addressed tools. Thus, even when we compare the SHC-pt results in complex scenario versus others in easier scenario the SHC-pt is better. CONCLUSIONS This paper presents two contributions. First, it proposes the method SentiHealth to detect the mood of cancer patients that are also users of communities of patients in online social networks. Second, it presents an instantiated tool from the method, called SentiHealth-Cancer (SHC-pt), dedicated to automatically analyze posts in communities of cancer patients, based on SentiHealth. This context-tailored tool outperformed other general-purpose sentiment analysis tools at least in the cancer context. This suggests that the SentiHealth method could be instantiated as other disease-based tools during future works, for instance SentiHealth-HIV, SentiHealth-Stroke and SentiHealth-Sclerosis.

A coevolutionary algorithm to automatic test case selection and mutant in Mutation Testing

One of the main problems to perform the Software Testing is to find a set of tests (subset from input domain of the problem) which is effective to detect the remaining bugs in the software. The Search-Based Software Testing (SBST) approach uses metaheuristics to find low cost set of tests with a high effectiveness to detect bugs. From several existing test criteria, Mutation Testing is considered quite promising to reveal bugs, despite its high computational cost, due to the great quantity of mutant programs generated. Therefore, this paper addresses the problem of selecting mutant programs and test cases in Mutation Testing context. To this end, it is proposed a Coevolutionary Genetic Algorithm (CGA) and the concept of Genetic Effectiveness, describing a new representation and implementing new genetic operators. The CGA is applied in five benchmarks and the results are compared to other five methods, showing a better performance of the proposed algorithm in subsets automatic selection with better mutation score and greater reduction of computational cost, specifically the amount of testing, when compared with exhaustive test.

An elitist evolutionary algorithm for automatically generating test data

The development of an effective and efficient method for generating test data is an extremely challenging process which directly impacts the time that could be spent on activities relevant to software testing. Therefore, various researches related to this area have been carried out. Among the techniques for automatically generating test data, we highlight the use of metaheuristics, a promising area called Search-Based Software Testing (SBST). Thus, this article proposes the use of an Elitist Genetic Algorithm (GA) as a tool for generation and selection of test data applied in Mutation Testing for different benchmarks. The results indicate a good performance of the algorithm used in the benchmarks.

Improved Crossover Operators for Genetic Programming for Program Repair

GenProg is a stochastic method based on genetic programming that presents promising results in automatic software repair via patch evolution. GenProg’s crossover operates on a patch representation composed of high-granularity edits that indivisibly comprise an edit operation, a faulty location, and a fix statement used in replacement or insertions. Recombination of such high-level minimal units limits the technique’s ability to effectively traverse and recombine the repair search spaces. In this work, we propose a reformulation of program repair operators such that they explicitly traverse three subspaces that underlie the search problem: Operator, Fault Space and Fix Space. We leverage this reformulation in the form of new crossover operators that faithfully respect this subspace division, improving search performance. Our experiments on 43 programs validate our insight, and show that the Unif1Space without memorization performed best, improving the fix rate by 34 %.

Shrinking a database to perform SQL mutation tests using an evolutionary algorithm

This paper tries to combine SQL mutation testing techniques with evolutionary computation aiming to improve the test data to SQL instructions. Based on a heuristic perspective it presents an approach that uses Genetic Algorithms (GA) to select tuples from an original database trying to reduce this one in an effective data set. The goal is to find a reduced data set which is able to detect a large number of faults in the SQL instructions of a given application. During the evolutionary process, the analysis of mutants is used to assess each set of data test selected by GA. The results obtained from experiments reveal a good performance using GA metaheuristic.

Improved representation and genetic operators for linear genetic programming for automated program repair

Genetic improvement for program repair can fix bugs or otherwise improve software via patch evolution. Consider GenProg, a prototypical technique of this type. GenProg’s crossover and mutation operators manipulate individuals represented as patches. A patch is composed of high-granularity edits that indivisibly comprise an edit operation, a faulty location, and a fix statement used in replacement or insertions. We observe that recombination and mutation of such high-level units limits the technique’s ability to effectively traverse and recombine the repair search spaces. We propose a reformulation of program repair representation, crossover, and mutation operators such that they explicitly traverse the three subspaces that underlie the search problem: the Operator, Fault and Fix Spaces. We provide experimental evidence validating our insight, showing that the operators provide considerable improvement over the baseline repair algorithm in terms of search success rate and efficiency. We also conduct a genotypic distance analysis over the various types of search, providing insight as to the influence of the operators on the program repair search problem.

SCOUT: A Multi-objective Method to Select Components in Designing Unit Testing

The creation of a suite of unit testing is preceded by the selection of which components (code units) should be tested. This selection is a significant challenge, usually made based on the team members experience or guided by defect prediction or fault localization models. We modeled the selection of components for unit testing with limited resources as a multi-objective problem, addressing two different objectives: maximizing benefits and minimizing testing cost. To measure the benefit of a component, we made use of metrics from static analysis (cost of future maintenance), dynamic analysis (risk of fault, and frequency of calls), and business value. We tackled gaps and challenges in the literature to formulate an effective method, the Selector of Software Components for Unit Testing (SCOUT). SCOUT provides an automated extraction of all necessary data followed by a multi-objective optimization process. SCOUT is a method able to assist testers in different domains, and the Android platform was chosen to perform our experiments, taking nine leading open-source applications as our subjects. SCOUT was compared with two of the most frequently used strategies in terms of efficacy. We also compared the effectiveness and efficiency of seven algorithms in solving a multi-objective component selection problem. Our experiments were performed under different scenarios, and reveal the potential of SCOUT in reducing the market vulnerability, compared to others approaches. To the best of our knowledge, SCOUT is the first method to assist in an automated way software testing managers in selecting components for the development of unit testing, combining static and dynamic metrics and business value.

A novel fitness function for automated program repair based on source code checkpoints

Software maintenance, especially bug fixing, is one of the most expensive problems in software practice. Bugs have global impact in terms of cost and time, and they also reflect negatively on a companys brand. GenProg is a method for Automated Program Repair based on an evolutionary approach. It aims to generate bug repairs without human intervention or a need for special instrumentation or source code annotations. Its canonical fitness function evaluates each variant as the weighted sum of the test cases that a modified program passes. However, it evaluates distinct individuals with the same fitness score (plateaus). We propose a fitness function that minimizes these plateaus using dynamic analysis to increase the granularity of the fitness information that can be gleaned from test case execution, increasing the diversity of the population, the number of repairs found (expressiveness), and the efficiency of the search. We evaluate the proposed fitness functions on two standard benchmarks for Automated Program Repair: IntroClass and ManyBugs. We find that our proposed fitness function minimizes plateaus, increases expressiveness, and the efficiency of the search.

A hybrid genetic algorithm for group formation at workplace

Corporate university (CO) become a trend in organizations. However, functional characteristics are little explored by CO for boosting knowledge creation. One of them is functional diversity that has been pointed as a factor to increase innovation, contributing for the development of clear strategies and quick responses to changes at workplace. On the other hand, an appropriate distribution of roles helps to promote individual responsibility and group cohesion. It also contributes to the strengthening of positive interdependence of group members. These aspects are fundamental to the development of cooperative work. In this work, we propose a model for group formation for learning approaching dichotomous functional roles and preferred roles in order to improve the group performance into the courses offered by CO at Court of Justice of Goiás (CJG). From the proposed model, we constructed three group formation algorithms. The first algorithm forms groups randomly (AA). The second is a canonical genetic algorithm (CGA). The third is a hybrid genetic algorithm (HGA). After, we performed a comparative analysis of the results reached by the three algorithms. We observed that the HGA achieves superior results than the CGA and AA.

O Teste de Mutação apoiado pelo Algoritmo Genético Coevolucionário com Classificação Genética Controlada

Este artigo situa-se no campo dos algoritmos geneticos coevolucionarios que objetivam a selecao de bons subconjuntos de casos de teste e mutantes, no contexto do Teste de Mutacao. Desse campo de estudo, selecionou-se e avaliou-se duas abordagens existentes. Tal avaliacao, subsidiou o desenvolvimento de um novo Algoritmo Coevolucionario com Classificacao Genetica Controlada (AGC − CGC). Para analisar a abordagem, 164 experimentos foram realizados comparando os resultados do algoritmo proposto com outros tres metodos aplicados em quatro benchmarks reais. Os resultados revelam uma melhora significativa do AGC − CGC sobre as outras abordagens quando se considera o aumento do escore de mutacao sem aumentar acentuadamente o tempo de execucao.