Carlos V. Paradis

E komo mai. Welcome.

You will find some of the projects and published work I am involved with on this page, some of which are authored on my previous name (Andrade Silva).

Most of my work revolves around remote collaboration with
different faculty in and outside the U.S. in several domains with structured and unstructured data acquisition and analysis.

Carlos V. Paradis resume

PhD Student in Computer Science

PhD Advisor: Dr. Susanne Still

G.A. Advisor: Dr. Rick Kazman

Groups: Machine Learning Lab and Data-driven Software Engineering Lab

University of Hawaii at Manoa (2016-?)

- MS Computer Science, University of Hawaii at Manoa (2014-2016)
- MS Software Engineering, Stevens Institute of Technology (Nov. 2013)
- B.Sc. Computer Science, Universidade Federal da Bahia, Brazil (Nov. 2013)

For Education, Awards and Honors please see my Resume . For an exhaustive list of my career activities please visit my LinkedIn Profile .

Ongoing Projects

Net-Zero Energy Building

Amid growing concerns about rising energy prices, energy independence, and the impact of climate change, statistics show buildings to be the primary energy consumer in the U.S. This fact underscores the importance of targeting building energy use as a key to decreasing the nation’s energy consumption.

One current solution to minimize building energy consumption are net-zero buildings. In the context of this project, a net-zero building is thought of as being capable to generate energy (from solar irradiance) and consume energy through it’s various energy devices. The goal at every year is minimally to be Net-Zero, as Net-Negative will inccur costs to purcharse energy. The energy management, however, is not fully automated by the building: It is the responsability of the building’s users to reach the Net-Zero goal every year.

Understanding key patterns in several types of devices through time and user behavior to identify actionable insight that minimizes energy consumption is the vision of this project.

In the ERDL Lab, my current tasks serve to assist in the creation of a dashboard for net-zero building residents. This is divided into preparing a database of sensor data, and performing sensor data analysis. Learned patterns are then encoded into visualizations that lead residents to make energy-saving decisions to achieve net-zero goals by the end of each calendar year.

An initial report of sensor data analysis of the project can be found on my Rpubs page.

Proactive Cybersecurity

The cybersecurity community typically reacts to attacks after they occur. Being reactive is costly and can be fatal where attacks threaten lives, important data, or mission success. But can cybersecurity be done proactively?

In this project, my group is collaborating with Dr Wang's group at the University of Maryland to instantiate a proactive approach to cybersecurity.

My role in the group consists of overseeing master students work, and investigate text mining algorithms that realize the project vision. More details will be provided on later stages of the project.

Solar Forecasting

(Master Thesis Plan A) Energy companies in Hawaii have started to use solar energy harvested in power plants as a more cost-effective alternative to provide energy to consumers. A problem with this renewable alternative is the dependence on weather, and consequent unpredictability of the next day solar irradiation.

Under the supervision of Dr. Lim and Dr. Stevens, I've been investigating how to identify such patterns using probabilistic models and information theory.

Some of our current results have been accepted on the International Conference on Machine Learning and Applications (ICMLA 2015).

Finished Projects and Published Work

A Tool to Enable Technology Transfer for Aerospace Concepts of Operation

Technical transfer of concepts of operation from NASA research areas into missions or to outside agencies like the FAA is difficult and expensive. The development of the research prototype does not follow the guidelines specified for operational software. As a result, the research prototypes cannot be transferred directly. The FAA, for example, instead asks NASA to deliver the requirements and specifications that an outside contractor will use to implement the final system. In order to meet certification requirements operational software systems are re-implemented following rigorous processes. The reverse-engineering of requirements and specifications from research prototypes is entirely manual and very costly. This is a significant hindrance in successful tech transfer.This proposal would automate key parts of the process of requirement and specification engineering and reduce the cost of technical transfer.

As a NASA Intern under Dr. Davies, I did a research survey on current practice of clone detection tools and used open source projects Klee, LLVM and Clang to instantiate a clone detection tool.

Fiscal Budget in Brazil

I volunteered as a data analyst for the Open Spending team of Open Knowledge Brazil Chapter , which was a finalist on Google Impact Challenge in Brazil .

The project's ambitious goal was to analyze the budgets of the city and states of São Paulo as well as the federal budget. These served as case studies to allow the method and tools to be applied to other cities.

My role in the team was to find, explore, pre-process, and analyze Brazil's Fiscal Year Budget. Across the promised outcomes, my role more closely related to helping craft stories, or by finding and exploring what data is available, help shape the presentation of it through the website and/or API.

This project served as a fine example of "parachuting" in on a project and quickly understanding the materials, identifying goals, and completing the required tasks for sucessful project continuity. Going beyond my individual tasks, I was often referred as the joker's card in the team because I summarized the project vision into a method diagram, and wrote the entire project roadmap.

The project is currently run by a third party organization.

Tactic Recommender

Together with Dr. Kazman and Dr. Mirakhorli , I've investigated how software engineers conduct web searching for a particular class of code snippets by (1) defining a suitable set of metrics to the problem space from literature review, (2) performing quantitative and qualitative data analysis to assess the chosen metrics usefulness for a software developer, (3) incorporate the set of metrics to a bigger set and evaluate prediction performance on recommending code snippets.

"How Toilets Makes People Taller"

World hunger is a problem usually brought to our attention on our everyday lives, however other factors may impact children malnutrition. Identifying such factors can help policy makers decide for the most appropriate intervention in order to improve people's life in underdeveloped countries.

Under the supervision of ScD. Fan I explored associations between anthropometric proxies of malnutrition (e.g. height, weight) and sanitation levels in sub-saharan african countries. This resulted in a full article and literature review presented as a class project in Bio Informatics during my MS in Computer Science. Both works received A.

Software Eng. Education

In general, computer programming courses are just focused on the language, constructions, structures, and so on. We worked on a new approach to teach an introductory computer programming course based on software engineering fundamentals. We emphasized during class project inter and intra-group software development, code review, issue reporting and other common software engineering practices. The approach was applied for 3 years in a university in Brazil and later published on the Brazilian Conference of Software Engineering (CBSoft 2011) .

Univ. Retention Patterns

In Public Universities in Brazil, undergraduate student retention, i.e. students graduating beyond the expected time in their majors, is a recurring problem in education. While the causes have been throughly studied on different countries, workable solutions for the university are seldom presented. This work identified students usually start falling behind on their freshman years. Furthermore, we hypothesized that the rearrangement of course workload could help reduce the retention problem after performing rule mining from student coursework choices throughout their student life. The work was later published on the Brazilian Symposium of Informatics and Education (SBIE 2013).

Software Eng. Effort Metrics

Identifying patterns of effort within software development projects is important for several reasons and a well-studied problem within the SE Community. One problem however is that companies often do not have available the necessary data or resources to make informed decision and hence benefit of current findings in the literature. For my honor thesis with Dr. Almeida and Dr. Kazman, I explored Open Source Projects for proxy measures of effort using Classic Grounded Theory and Data Exploration methods. Part of this work appeared in the Economics-Driven Software Architecture Journal 2014.

Code Understanding and Navigation

The concept of code smells is widespread in Software Engineering. However, in spite of the many discussions and claims about them, there are few empirical studies to support or contest these ideas. In particular, the study of the human perception of what is a code smell and how to deal with it has been mostly neglected. We performed a controlled experiment that extends and builds upon another empirical study about how humans detect god classes. Our findings published on the International Conference on Evaluation and Assessment in Software Engineering (EASE 2013) show that participants have different personal criteria and preferences in choosing drivers to identify god classes.

Publications

SILVA, Carlos V. A. ; LIM, Lipyeow ; STEVENS, Duane ; NAKAFUJI, Dora. Probabilistic Models for One-Day Ahead Solar Irradiance Forecasting in Renewable Energy Applications. Special Track on Machine Learning on Energy Applications. Internacional Conference on Machine Learning and Applications 2015.

Martin Naedele, Hong-Mei Chen, Rick Kazman, Yuanfang Cai, Lu Xiao, Carlos V.A. Silva , Manufacturing execution systems: A vision for managing software development, Journal of Systems and Software, Volume 101, March 2015, Pages 59-68, ISSN 0164-1212, http://dx.doi.org/10.1016/j.jss.2014.11.015.

SILVA, C. V. A. ; SANTOS, M. S. ; Claro, D.B. ; Silva, Veronica ; SILVA, M. ; RIBEIRO, S. ; TELLES, A. R. ; LOPES, D. . Mining Retention Rules from Student Transcripts: A Case Study of the Information Systems programme at a Federal University. Anais do Simpósio Brasileiro de Informática na Educação, v. 1, p. 1, 2013.

ANTOS, JOSÉ A. M. ; DE MENDONÇA, MANOEL G. ; SILVA, CARLOS V. A. . An exploratory study to investigate the impact of conceptualization in god class detection. In: the 17th International Conference, 2013, Porto de Galinhas. Proceedings of the 17th International Conference on Evaluation and Assessment in Software Engineering - EASE '13. New York: ACM Press, 2013. p. 48.

ALMEIDA, E.S ; MACHADO, I. C. ; SILVA, C. V. A. ; GOMES, G. S. S. . Teaching Software Engineering Fundamentals in an Introductory Computer Programming Course. In: IV Fórum de Educação em Engenharia de Software (FEES), 2011, São Paulo. XXV Simposio Brasileiro de Engenharia de Software (SBES), 2011.

U. Federal da Bahia

At Universidade Federal da Bahia, Brazil, I taught Databases Lab, as well as Data Structures and Information Retrieval to Information Systems students. I used a combination of Latex, "hand-made" Tikz forms, color, and other visual examples to ilustrate abstract concepts for freshman, together with lecture notes.

Quizes, and projects often followed familiar concepts related to gaming and social media, while coding assignments evolved in difficulty from filling in the blank to entire group work projects from scratch and cross-group evaluation.

Assistant Professor

You can find the Data Structure class material on my Dropbox, both slides and notes in latex and pdf.

ACM Honors Society - UPE

Upsilon Pi Epsilon (UPE) is an honorary society whose membership consists of outstanding undergraduate and graduate students in Computing and Information Disciplines. Members are chosen not only for their scholastic achievement in a computing science program, but also for distinguishing themselves as true professionals by meeting the standards of the society such as outstanding talent, high scholarship, high standards, representation of computing science in interdisciplinary communication, and encouragement of individual contributions to society through computing science.

After induction at Stevens Institute of Technology chapter in 2012, I collaborated with the group with local workshops for students and job interview preparations.

IEEE Honors Society - HKN

Eta Kappa Nu (HKN) is the National Honor Society for Electrical and Computer Engineers. Members consist of students, alumni, and other professionals who have demonstrated exceptional academic and professional accomplishments. Student members are selected on the basis of scholastic standing, an impeachable character, and excellent leadership skills.

While participating as a Pledge in 2015-2016 of this chapter, I engaged on various activities to Hawai'i community from volunteering as a teacher assistant for EE undergraduates to community service projects.

Science Without Borders Network (Rede CsF)

Rede CsF is a non-profit organization based in Brazil and created by students who were awarded the Science Without Border scholarship from Brazil government, which started in 2012 and I was among the first students.

This grant hosts you a full year (board,housing,health insurance, tuition and a small stipend) abroad at one of the eligible countries. A student investment is very high, but the only return required by the program is for a student to stay in Brazil for the equivalent amount of time. Some of the awarded students, like me, feel that this is not a realistic return to the country for the high investment. This is the purpose of Rede CsF: To connect awarded students by the Science Without Border program to enable projects with anyone interested to perform projects for the country. The focus of the network is on Science, Technology, Innovation and Education.

With more than 1890 members and 40 contributors making us the largest network of awarded students of the country, our group was recently featured on the government program website.

Within Rede CsF I contribute mostly with data analytics across our online platforms (Website, Facebook Pages and Groups, WhatsApp, Youtube and Trello) and managerial tasks within the network. For instance, I created the Intranet of Rede CsF on Google Apps to help with collaboration of ideas, and share of feedback.