The expansion of Data Science projects in organizations has been led by three factors: the growth in the amount of data generated, the evolution in storage capacity, and the increase in computational capabilities. However, most of these projects fail to deliver the expected value: 82% of the teams do not use any process model. Despite the popularity of Agile Methods, their adoption in Data Science projects is still scarce. Most of the existing research focuses on algorithms. There is a lack of studies on agility in Data Science. This Systematic Literature Review (SLR) was performed to identify and evaluate 16 studies that can answer how to adapt and apply CRISP-DM using different approaches — methods, frameworks, or process models. In addition, it shows how CRISP-DM has evolved over the last few decades, with derivations emerging from rigid processes to agile methods. This research then analyzes the 16 tailored models and examines the similarities and differences between CRISP-DM derivatives. As a result, it summarizes the CRISP-DM adaptation patters identified, such as phase addition, phase modification, features and tools addition, and integration with other approaches. Consequently, this SLR showcases how CRISP-DM is a robust, flexible, and highly adaptable model that can be extended to different business domains. Finally, it proposes a theoretical guide to modify and customize CRISP-DM for Data Science projects.
Text-based social media channels, such as Twitter, produce torrents of opinionated data about the most diverse topics and entities. The analysis of such data (aka. sentiment analysis) is quickly becoming a key feature in recommender systems and search engines. A prominent approach to sentiment analysis is based on the application of classification techniques, that is, content is classified according to the attitude of the writer. A major challenge, however, is that Twitter follows the data stream model, and thus classifiers must operate with limited resources, including labeled data and time for building classification models. Also challenging is the fact that sentiment distribution may change as the stream evolves. In this paper we address these challenges by proposing algorithms that select relevant training instances at each time step, so that training sets are kept small while providing to the classifier the capabilities to suit itself to, and to recover itself from, different types of sentiment drifts. Simultaneously providing capabilities to the classifier, however, is a conflicting-objective problem, and our proposed algorithms employ basic notions of Economics in order to balance both capabilities. We performed the analysis of events that reverberated on Twitter, and the comparison against the state-of-the-art reveals improvements both in terms of error reduction (up to 14%) and reduction of training resources (by orders of magnitude).
Respiratory insufficiency is a medic symptom in which a person gets a reduced amount of oxygen in the blood. This paper reports the experience of building SPIRA: an intelligent system for detecting respiratory insufficiency from voice. It compiles challenges faced in two succeeding implementations of the same architecture, summarizing lessons learned on data collection, training, and inference for future projects in similar systems.
Applying agile practices in data science requires adaptations. This paper describes challenges and lessons learned in two applied machine learning projects developed in the XP Lab course at University of São Paulo in Brazil. It compiles six suggestions for educators and practitioners who want to bring agility to their data science initiatives.