No abstract

Quite recently, considerable attention has been paid to developingartificial intelligence and data science areas. This has been drivenby scientific advances and the growing number of software andservices that are popularizing machine learning techniques andalgorithms and driving people with less knowledge in areas suchas statistics and mathematics to create their predictive models. Asa result, the machine learning field is no longer only scientificand has aroused the interest of companies from different domains.These events led to the emergence of multiple tools such as Scikit-Learn, Tensorflow, Keras, Pycaret, and a vast number of cloud-basedmachine learning services that provide an acceleration in the developmentof predictive models at speeds never seen. However, manychallenges remain in operationalizing and maintaining machinelearning-centered products, making many business initiatives frustrated.In this scenario, practical experience shows that machinelearning is only a slice of a more extensive set of practices andtechnologies necessary to build solutions in this area. In this paper,the main goal is to identify the challenges currently faced by datascientists in developing Machine Learning-centric products andhow Machine Learning Operations can support overcoming them.For this purpose, a survey was conducted that collected answersfrom 66 Brazilian professionals in data science. From the challengesidentified, the importance of Machine Learning Operations practicesas an integrated part of the Machine Learning lifecycle wasexplored. Finally, this work contributes to filling the gap in MachineLearning Operations in daily activities involving data science andadvancing this research field in Brazil.

It can be challenging for people to select the most relevant requirementamong several software system development options.Requirements prioritization defines the ordering for executing requirementsbased on their priority or importance concerning stakeholders’viewpoints, which is a problematic task. Based on thisproblem, this study aims to present a requirements prioritizationapproach using a genetic algorithm to find optimal solutions, andit can assist in the requirements prioritization activity during thesoftware development process. In this paper, we investigated aset of criteria to create four functions GUT-D, ThS-D, ST, and LT,to assess candidate solutions, i.e., the recommended prioritizedrequirements. We examine the empirical results concerning thepractical approach’s effectiveness and cost computational two experimentsin the evaluation. We found that the 𝐺𝑈𝑇 − 𝐷 fitnessfunction achieved the best fitness value in different settings with90.51% and 98.63%. Besides that, our study demonstrates that the approachis promising to assist requirements prioritization since eachfitness function can be used individually according to companies’necessities.

Students with visual impairment need for different methods to assist their studies such as read tools and conversation bots (chatbots). Assistive technologies can be incorporated as a training or support tool and specifically in the educational field as complementary material. This paper proposes a chatbot as a study that supports an intelligent system for visually impaired students. Finally, the paper reports a pilot experiment regarding the effectiveness, time of response, and speech output of the proposed system in two scenarios of the software engineering area. Overall, the system proved to be tolerant of typing errors and returned quick responses by 0,9 seconds on average. Our results also demonstrate the efficiency in output with speech employing a library called gTTS.

A depressão é um distúrbio psicológico que afeta milhões de pessoas no mundo, indiferente à idade, classe social ou nacionalidade. Diferentes técnicas tem sido exploradas para analisar e reconhecer sintomas depressivos na literatura, como técnicas de Processamento de Linguagem Natural e Análise de Sentimentos. Entretanto, para o português brasileiro, poucos estudos tem proposto datasets para a classificação de sintomas da depressão. Neste artigo, propomos uma estratégia chamada DP-Symptom-Identifier para coletar tweets e criar um novo dataset com sentenças que possuem sintomas da depressão. Experimentos iniciais usando diferentes algoritmos obtiveram um alto desempenho preditivo, o que mostra que as pesquisas nesta área são promissoras.