Federated Neuromorphic Learning (FNL) applies Spiking Neural Networks (SNNs) to enable energy-efficient collaborative learning on devices without centralizing data.However, integrating Differential Privacy (DP) introduces critical changes to the SNN firing dynamics, which propagate to server coordination strategies.This paper investigates DPinduced firing-rate distortions and their influence on global model convergence and generalization.Experimental ablation studies across privacy budgets and clipping bounds highlight firing distortions directly related to global accuracy degradation.Additionally, client selection instabilities related to DP noise degrade the model aggregation performance.The results reinforce that firing-rate-based FNL strategies are fragile under DP and require precise calibration to maintain the effectiveness of federated coordination.

Abstract Scrum is widely adopted in software project management due to its adaptability and collaborative nature. The recent emergence of Large Language Models (LLMs) has created new opportunities to support knowledge-intensive Scrum practices. However, existing research has largely focused on technical activities such as coding and testing, with limited evidence on the use of LLMs in management-related Scrum activities. In this study, we investigate the use of LLMs in Scrum management activities through a survey of 70 Brazilian professionals. Among them, 49 actively use Scrum, and 33 reported using LLM-based assistants in their Scrum practices. The results indicate a high level of proficiency and frequent use of LLMs, with 85% of respondents reporting intermediate or advanced proficiency and 52% using them daily. LLM use concentrates on exploring Scrum practices, with artifacts and events receiving targeted yet uneven support, whereas broader management tasks appear to be adopted more cautiously. The main benefits include increased productivity (78%) and reduced manual effort (75%). However, several critical risks remain, as respondents report ‘almost correct’ outputs (81%), confidentiality concerns (63%), and hallucinations during use (59%). This work provides one of the first empirical characterizations of LLM use in Scrum management, identifying current practices, quantifying benefits and risks, and outlining directions for responsible adoption and integration in Agile environments.

Abstract Background: Ensuring the quality of user stories is vital to Agile Software Development. Rule-based tools like AQUSA, based on the Quality User Story (QUS) framework, offer reliable structural checks but struggle with context-sensitive or pragmatic issues. Large Language Models (LLMs) have emerged as potential alternatives, yet prior studies often rely on small datasets, older models, or lack direct comparison with rule-based baselines. Objective: This study aims to assess the effectiveness of modern LLMs relative to a rule-based tool (AQUSA) for detecting defects in user stories, considering both structural and contextual dimensions. Method: We conduct a large-scale comparative evaluation involving AQUSA and three GPT-family LLMs (GPT-5, GPT-5-mini, and GPT-4), using 182 user stories drawn from three industrial datasets. We apply both quantitative metrics (precision, recall, F1-score) and qualitative analysis of feedback clarity and defect relevance. Results: GPT-5-mini achieved the highest recall (0.81) and overall F1-score (0.62), while AQUSA attained the highest precision (0.61) with significantly fewer false positives. GPT-5 showed high hallucination rates and instability; GPT-4 was overly conservative, leading to under-detection of defects. Conclusion: Neither rule-based nor GPT-family LLM-based approaches suffice in isolation. Rule-based tools enforce structural rigor, while LLMs capture nuanced linguistic and pragmatic flaws. We advocate a hybrid “Dual-gate” strategy—using AQUSA for structural validation followed by lightweight LLMs for contextual refinement—to improve the reliability and scalability of user story quality assessment in agile environments.

Technical Debt (TD) is a metaphor reflecting technical compromises that can yield short-term benefits but might hurt the long-term health of a software system. With the increasing amount of data generated when performing software development activities, an emergent research field has gained attention: applying Intelligent Techniques to solve Software Engineering problems. Intelligent Techniques were used to explore data for knowledge discovery, reasoning, learning, planning, perception, or supporting decision-making. Although these techniques can be promising, there is no structured understanding related to their application to support Technical Debt Management (TDM) activities. Within this context, this study aims to investigate to what extent the literature has proposed and evaluated solutions based on Intelligent Techniques to support TDM activities. To this end, we performed a Systematic Mapping Study (SMS) to investigate to what extent the literature has proposed and evaluated solutions based on Intelligent Techniques to support TDM activities. In total, 150 primary studies were identified and analyzed, dated from 2012 to 2021. The results indicated a growing interest in applying Intelligent Techniques to support TDM activities, the most used: Machine Learning and Reasoning under uncertainty. Intelligent Techniques aimed to assist mainly TDM activities related to identification, measurement, and monitoring. Design TD, Code TD, and Architectural TD are the TD types in the spotlight. Most studies were categorized at automation levels 1 and 2, meaning that existing approaches still require substantial human intervention. Symbolists and Analogizers are levels of explanation presented by most Intelligent Techniques, implying that these solutions conclude a general truth after considering a sufficient number of particular cases. Moreover, we also cataloged the empirical research types, contributions, and validation strategies described in primary studies. Based on our findings, we argue that there is still room to improve the use of Intelligent Techniques to support TDM activities. The open issues that emerged from this study can represent future opportunities for practitioners and researchers.

Technical Debt (TD) refers to the consequences of taking shortcuts when developing software. Technical Debt Management (TDM) becomes complex since it relies on a decision process based on multiple and heterogeneous data, which are not straightforward to be synthesized. In this context, there is a promising opportunity to use Intelligent Techniques to support TDM activities since these techniques explore data for knowledge discovery, reasoning, learning, or supporting decision-making. Although these techniques can be used for improving TDM activities, there is no empirical study exploring this research area. This study aims to identify and analyze solutions based on Intelligent Techniques employed to support TDM activities. A Systematic Mapping Study was performed, covering publications between 2010 and 2020. From 2276 extracted studies, we selected 111 unique studies. We found a positive trend in applying Intelligent Techniques to support TDM activities, being Machine Learning, Reasoning Under Uncertainty, and Natural Language Processing the most recurrent ones. Identification, measurement, and monitoring were the more recurrent TDM activities, whereas Design, Code, and Architectural were the most frequently investigated TD types. Although the research area is up-and-coming, it is still in its infancy, and this study provides a baseline for future research.

Agile Software Development (ASD) has become the mainstream software development method of choice. Its core fundamentals are based on Teamwork factors and the higher value of individuals and their interactions over processes and tools. However, there is no common understanding regarding the factors that should be considered for defining an ASD Teamwork construct. Driven by this problem, we present a thematic network that synthesizes the information presented in the literature, and eases knowledge sharing by defining a terminology. The thematic network is the result of the following process: (i) studies definition to be used as data source through a literature review; (ii) data extraction from these studies; (iii) data translation into codes; (iv) codes translation into themes; (v) creation of higher-order themes model; and (vi) assessment of synthesis trustworthiness. The resulting thematic network comprises four higher-order themes: Cohesion, Orientation, Shared Leadership, and Autonomy. We also evaluate the applicability of the identified themes in ASD Teamwork constructs in the literature. We concluded that the constructed thematic network can be generalized to ASD, and used as basis by researchers who intend to explore ASD Teamwork. Further, practitioners can use our results to understand agile teams’ dynamics better and improve their efficiency.

Agile Software Development (ASD) has become the mainstream software development method of choice.Its core fundamentals are based on Teamwork factors and the higher value it gives to individuals and their interactions over processes and tools.Teamwork and human factors have been addressed as essential topics in the literature, and researchers have stated the importance of measuring it to increase the chances of success of ASD projects.However, there is no common understanding regarding the factors that should be considered for defining an ASD Teamwork construct.Driven by this problem, this paper presents a thematic network that defines the themes (i.e., factors) that should be considered when addressing ASD Teamwork.The ASD Teamwork thematic network is the result of a process that consisted of (i) defining the studies used as a data source through a literature review; (ii) extracting data from these studies; (iii) translating this data into codes; (iv) translating the codes into themes; (v) creating the model of higher-order themes; and, (vi) assessing the trustworthiness of the synthesis.The resulting thematic network comprises four higher-level themes: Cohesion, Orientation, Shared Leadership, and Autonomy.We believe that the constructed thematic network can be generalized to ASD and used as the basis by researchers who intend to explore ASD Teamwork.Further, practitioners can use our results to understand agile teams' dynamics better and improve their efficiency.

Over the years, researchers have explored the influence of human factors in software engineering, showing that the team members' personalities might affect teamwork.However, it is challenging to measure software engineers' personalities due to the number of available psychometric instruments and the possibility of using different scales and classifications.Our study compares the personality traits measured by three psychometric instruments used in Software Engineering: Big Five Inventory (BFI), 16 Personality Factors (16PF), and Context Cards (CC).For this purpose, we executed an empirical study in which we collected data from 29 software developers for each of the evaluated instruments.As a result, we identified a moderate correlation between BFI and 16PF, confirming the current stateof-the-art.For the remaining combinations, there was a weak correlation.As implications for this research, there is a need to empirically evaluate BFI and CC (context-specific survey) in terms of construct validity since they have moderate to low correlation.

Using knowledge-based systems for helping agile teams to improve their performance is not a fact in the industry.In previous work, we have presented Kaizen, a knowledge-based Bayesian network for assisting Scrum teams in diagnosing their value stream in light of the predicted Customer Satisfaction and, consequently, improve their performance.This study assesses Kaizen's accuracy to predict Customer Satisfaction using realworld data.We adopted Kaizen for one software development company and collected data from 18 projects using an online questionnaire.We collected two types of data: inputs for Kaizen and the expected Customer satisfaction.We used the first type of collected data as inputs for Kaizen to calculate the predicted Customer satisfaction.Then, we assessed Kaizen's accuracy by comparing the predicted (i.e., calculated) and expected (i.e., collected) Customer satisfaction using face value and the average Brier score.Considering the face value, Kaizen predicted Customer Satisfaction correctly for 14 out of the 18 projects.The average Brier Score was 0.16.The model predicts, with satisfactory accuracy, the Customer Satisfaction and systemizes the process for Scrum teams to self-diagnose, enabling for causal analysis and supporting their continuous improvement.

Accurate effort estimation is an important part of the software process.In Agile Software Development, the techniques for predicting effort are mostly based on expert judgment, but there are approaches based on Machine Learning.The theme continues to be challenging and a subject of further studies given the difficulty of finding accurate solutions to the problem.This paper proposes and evaluates a tool based on the decision tree method for effort estimation in agile projects.We evaluated our tool given its accuracy and ease of use collecting data from four projects.To evaluate the accuracy, we compared the values of Magnitude of Relative Error from the teams' estimations with the values provided by the tool.To evaluate the ease of use, we used the Technology Acceptance Mode.The initial results show that the tool can be reliably used with minimal training.In terms of accuracy, the tool achieved lower error compared to the estimates provided by the teams (mean: 19.05% vs 33.32%), and the evaluation means in TAM were higher than 4.0 in ten of the eleven variables analyzed on a Likert scale.From this work, we conclude that estimation by decision tree is a viable technique that, at the very least, can be used by project managers to complement current estimation techniques.

One challenge in constructing a Bayesian network (BN) is defining the node probability tables (NPTs), which can be learned from data or elicited from domain experts.In practice, for large-scale BN it is common not to have enough data for learning and elicitation from experts is unfeasible.Previous work proposed a solution to this problem: the Ranked Nodes Method (RNM).However, this solution needs to be applied by a RNM expert who, through the elicitation of expert judgement, identifies the necessary parameters for the RNM algorithm to generate the NPTs.Hence, this paper presents a novel approach to define NPT using the RNM with no ranked nodes-specific knowledge.The solution is named Simulated Bayesian Network Expert (SBNE).It consists of eliciting a subset of the NPT from the domain experts which is used as input to an algorithm that estimates the optimal parameters for the RNM to generate the NPTs.To validate our solution, we conducted an experiment with multiple domain experts and compared the results with other methods.Our solution outperformed the other methods (producing NPTs at least 12% more accurate) and is, therefore, a promising approach to apply RNM without relying on RNM experts.

Due to the need for flexibility to requirements changes, agile software development methods have been attracting the attention of academic and industrial domains.Unlike traditional approaches, agile methods focus on the rapid delivery of business value to customers through empirical and incremental development processes.Despite being effective in delivering quality functional requirements, agile practices generally neglect non-functional requirements until the later stages of software development.However, neglecting non-functional requirements during requirements analysis can lead to project failures.In this paper, we present the NFRec tool, which aims to support software developers in the elicitation of non-functional requirements in the context of agile software development.Additionally, we report the results from a case study to evaluate the acceptance of the NFRec tool from the point of view of software developers of four projects from a Brazilian software company.To gather information about the tool acceptance, we applied a questionnaire based on the indicators from the Technology Acceptance Model.Overall, the four teams considered the NFRec tool useful and easy to use for supporting the management of non-functional requirements in agile projects.

Software metrics are usually used for quantification, not giving the necessary support for decision making.To increase their usefulness, it is necessary to give them meaning through the definition of significant thresholds.Despite its importance, the state of the art on threshold derivation is mostly based on data-driven approaches.This paper presents a systematic approach to define thresholds for metrics in the absence of data and based on eliciting knowledge from experts.The proposed approach is based on identifying context factors that influence the thresholds for a given metric and is supported by fuzzy logic concepts to model the crisp value (i.e., collected data) into a linguistic variable (i.e., interpreted information).We present context factors elicited from three experts for the metrics code coverage, static code analysis warnings count and defect count.Further, we present cases on how to implement the proposed approach.As a result, we conclude that the approach is promising.

Context: Companies must make a paradigm shift in which both short- and long-term value aspects are employed to guide their decision-making. Such need is pressing in innovative industries, such as ICT, and is the core of Value-based Software Engineering (VBSE). Objective: This paper details three case studies where value estimation models using Bayesian Network (BN) were built and validated. These estimation models were based upon value-based decisions made by key stakeholders in the contexts of feature selection, test cases execution prioritization, and user interfaces design selection. Methods: All three case studies were carried out according to a Framework called VALUE — improVing decision-mAking reLating to software-intensive prodUcts and sErvices development. This framework includes a mixed-methods approach, comprising several steps to build and validate company-specific value estimation models. Such a building process uses as input data key stakeholders’ decisions (gathered using the Value tool), plus additional input from key stakeholders. Results: Three value estimation BN models were built and validated, and the feedback received from the participating stakeholders was very positive. Conclusions: We detail the building and validation of three value estimation BN models, using a combination of data from past decision-making meetings and also input from key stakeholders.

The swarm concept in Internet of Things (IoT) describes the cooperation of independent and heterogeneous devices to execute tasks. Components of a swarm system must seamlessly discover other objects. Semantic discovery is an approach to perform this task and can be performed either by automated or manual ways, but not often taking account information trust. This article presents a middleware that performs automatic IoT semantic discovery and determining information trust. Experimental results, carried out with real data from smart cities projects are presented.

The software industry's current decision-making relating to product/project management and development is largely done in a value neutral setting, in which cost is the primary driver for every decision taken. However, numerous studies have shown that the primary critical success factor that differentiates successful products/projects from failed ones lie in the value domain. Therefore, to remain competitive, innovative and to grow, companies must change from cost-based to value-based decisionmaking where the decisions taken are the best for that company's overall value creation. This paper details a case study where value-based decisions made by key stakeholders to select features for the next sprint of an Internet of Things (IoT) project, stored in a decisions database, were used to build and validate a value estimation model. This model's goal was to estimate the overall value contribution that each feature being discussed during a decision-making meeting would bring to the company, if selected for implementation. The estimation technique employed was Bayesian Network, and validation results were quite positive.

Search-Based software engineering (SBSE) deals with metaheuristic search-based optimization techniques to provide solutions for complex problems.A popular problem in literature is the team formation problem (TFP), which consists of finding the best allocation of human resources to a software development project.This problem is recognized as NP-hard and it is more complex in companies that carry out multiple projects.This paper presents an effective and automated approach to allocate multiple developers into multiple teams to maximize the technical compatibility between them.The approach consists of an SBSE method that uses Genetic Algorithm to simultaneously build multiple teams, using data from tag-based profiles.We conducted an empirical evaluation using data from eight realworld software projects of a Brazilian company.The results indicate that tag-based profiles is a promising information source to represent technical knowledge, since the suggested teams were considered to have the proper skills to the attend the technical demand of the projects.The approach was able to reach high levels of satisfaction, delivering teams in an effective and automated way.Although, further investigation needs to be conducted to reach stronger conclusions.

One of the main issues of an agile software project is how to accurately estimate development effort.In 2014, it was published a Systematic Literature Review (SLR) regarding this subject.The authors of this SLR analyzed works from 2001 to 2013 and reached the number of 25 relevant papers.Therefore, the goal of our work is to provide an updated review of the state of the art based on this reference SLR work.We applied a Forward Snowballing approach, in which our seed set are the former SLR and its selected papers.We identified changes in this new review comparing it with the reference SLR: XP methodology was mentioned in just a few works; Use Case Points (UCP) method and Case Points as size metric were not found.We also observed a strong indication of solutions based on Artificial Intelligence and Machine Learning methods for effort estimation in Agile Software Development (ASD).Finally, we identified that in the reference SLR there is a gap in terms of agreement on suitable cost drivers.Thus, in our updated review, we applied Thematic Analysis in the selected papers and identified a representative set of 10 cost drivers for effort estimation.

Background: The focus of Agile software development (ASD) is different than plan-driven development, requiring new software process improvement (SPI) paradigms.Objective: To identify and synthesize the possible gaps of Agile improvement solutions (AIS) given their focus on people factors, report of successful adoption in industry projects and availability of tool support.Method: We applied a Systematic Literature Review of studies published up to (and including) 2017 through backward and forward snowballing given a start set.Results: In total, we evaluated 55 papers, of which 44 included AIS and the main findings are: 1) 26 consider teamwork factors; 2) 21 were applied on industry; 3) 10 out of these 21 presented evidence of increase in company performance; and 4) 19 of the solutions are for the purpose of adoption, 18 for assessment and 8 are maturity models.Conclusion: The main implication for this research is a need for more and better empirical studies documenting and evaluating AIS.For the industry, the review provides a map of current AIS approaches and can be used as a starting point to adopt agile SPI.

Agile software development focuses on quick delivery and flexibility to change.Despite being effective in delivering quality functional requirements, agile practices tend to neglect non-functional requirements until the later stages of software development.This work focuses on Scrum, the most popular agile method, and presents a non-functional requirements recommendation system to support Scrum practitioners on their early identification.The solution is based on instrumenting the Scrum process to extract useful data and the use of collaborative filtering and item recommendation.To evaluate the recommendations, we conducted off-line experiments with data collected from 12 Scrum practitioners through a survey.The data was analyzed using 10fold cross-validation.As a result, our proposed solution showed a recall rate of up to 81%, which indicates that it is a promising approach to recommend non-functional requirements given a set of functional requirements identified by project stakeholders.

One of the main issues of an agile software project is how to accurately estimate development effort. In 2014, a Systematic Literature Review (SLR) regarding this subject was published. The authors concluded that there were several gaps in the literature, such as the low level of accuracy of the techniques and little consensus on appropriate cost drivers. The goal of our work is to provide an updated review of the state of the art based on this reference SLR work. We applied a Forward Snowballing approach, in which our seed set included the former SLR and its selected papers. We identified a strong indication of solutions based on Artificial Intelligence and Machine Learning methods for effort estimation in Agile Software Development (ASD). We also identified that there is a gap in terms of agreement on suitable cost drivers. Thus, we applied Thematic Analysis in the selected papers and identified a representative set of 10 cost drivers for effort estimation. This updated review of the state of the art resulted in 24 new relevant papers selected.

Internet of Things (IoT) resources cooperate with themselves for requesting and providing services. In heterogeneous and complex environments, those resources must trust each other. On-Off attacks threaten the IoT trust security through nodes performing good and bad behaviors randomly, to avoid being rated as a menace. Some countermeasures demand prior levels of trust knowledge and time to classify a node behavior. In some cases, a malfunctioning node can be mismatched as an attacker. In this paper, we introduce a smart trust management method, based on machine learning and an elastic slide window technique that automatically assesses the IoT resource trust, evaluating service provider attributes. In simulated and real-world data, this method was able to identify On-Off attackers and fault nodes with a precision up to 96% and low time consumption.

[Context]: To remain competitive, innovative and to grow, companies should use a value-based decision-making where decisions are the best for that company's overall value creation. However, without tool support, the use of explicit value propositions and aggregation of different key stakeholders' decisions during decision-making may be a challenge for many companies. [Goal]: The goal of this paper is to investigate the extent to which a Web-based tool for value-based decision-making can successfully support stakeholders' decision-making process. [Method]: We conducted three case studies across four software projects, during six weeks, in the contexts of feature selection, test cases execution prioritization and user interfaces design selection. Prior to using the tool, stakeholders' value propositions were elicited via focus-group meetings; later, during a post-mortem phase, data was gathered via observation, semi-structured interviews and structured questionnaires. [Results]: Participants reported an improvement of their decision-making process and quality of decisions; further, they also felt confident about using the tool, and that it can be useful to their work. [Conclusions]: Results suggested that the use of tool support by the stakeholders in the investigated company for value-based decision-making improved their decision-making process and the quality of decisions.

Background: Definition of Done (DoD) is a Scrum practice that consists of a simple list of criteria that adds verifiable or demonstrable value to the product. It is one of the most popular agile practices and assures a balance between short-term delivery of features and long-term product quality, but little is known of its actual use in Agile teams.

Agile software development has been increasingly used to satisfy the need to respond to fast moving market demand and gain market share.Scrum, which is a project management framework, dominates as the most popular agile method.In the literature, there are a number of solutions to customize and assess Scrum-based agile methods, but they are limited to focus only on process factors, assume a predefined set of practices or rely only on subjective evaluation.This paper presents a framework to build a Bayesian Network to assist on the assessment of Scrum-based software development methods.The BN models the main entities of the software development process and can be complemented with software practices and metrics.To evaluate the completeness of our solution, we performed simulations to check if the proposed framework diagnoses 14 known Scrum anti-patterns extracted from the literature.12 antipatterns were directly detected, 1 was indirectly detected by the BN and 1 was considered as invalid.We concluded that the proposed solution is complete to detect the major flaws of Scrum-based software development methods and can be used to assist on the configuration, adoption and continuous improvement of Scrum teams.

Software metrics have a fundamental role in the process of software quality management.However, in most cases, they are only used to quantify attributes, not supporting decision-making during the software life cycle.To support decision-making, it is necessary to give them by defining thresholds.In the literature, several approaches have been proposed with this purpose.On the other hand, most of them do not consider context factors such as the domain.Given this, in this paper, we evaluate if context factors influence the definition of thresholds for software metrics.Our work is restricted to Chidamber and Kemerer metrics, due to availability of data.We conducted an empirical study composed of two quasi-experiments.Each quasi-experiment uses an approach presented in the literature to define thresholds for software metrics, with the defined thresholds as the dependent variable.As the factor, we used a variable with two possible treatments: to consider the context or not.To define context, we used factors presented in the literature.As the objects of study, we used the source code of fifteen Java-based open-source projects.For measurement purposes, we used the six original Chidamber and Kemerer metrics.For both quasi-experiments, the accuracy of the definition of thresholds improved by considering the context.Therefore, we concluded that context factors influence the definition of the threshold for Chidamber and Kemerer metrics, which is an indicator that it influences other software metrics.

Software metrics are essential resources in software enterprises.They can be used to support decision-making and, consequently, reduce costs, improve the productivity of the team and the quality of products delivered.On the other hand, this is only possible if the metrics are valid.Although there are studies related to software metrics validity, none present a solution to represent the uncertainties of the metrics selected to measure the attributes of the entities.In this paper, we present a process to build Bayesian networks to represent the uncertainties of software metrics-based models.The proposed solution is composed of two activities and focuses on the selection and validation of metrics to construct the Bayesian networks.We validated the model with simulated scenarios.Given the successful results, we concluded that the proposed solution is promising.This paper complements the state of the art by showing how to complement a popular metric selection technique, GQM, with information to model uncertainties of the metrics using the concepts of metric validation and Bayesian networks.

Abstract Scrumis a simple process to understand, but hard to adopt. Therefore, there is a need for resources to assist on its adoption. In this paper, we present the process followed to build aBayesian networkto assist on the assessment of the quality of the software process in the context ofScrumprojects. The model provides data to helpScrum Masterslead the improvement of business value delivery ofScrumteams. The process is divided into 2 phases. In the first phase, we built theBayesian networkbased on expert knowledge extracted from the literature and experts. We used a top‐down approach and reasoning to define the key metrics necessary to build the models and their relationships. In the second phase, we updated theBayesian networkbased on limitations of the first version. We validated theBayesian networkinferences with 10 simulated scenarios. Comparing both versions, for all scenarios, we improved the accuracy of the inferences. Therefore, we concluded that theBayesian networksadequately representScrumprojects from the viewpoint of theScrumMaster. Finally, the model built is in conformance with agile methods tailoring and can be adapted to anyScrumteam.

A research work is reproducible when all research artifacts such as as text, data, figure and code are available for independent researchers reproduce the results. In this paper, we present a reproducible gait analysis to track Parkinson's Disease evolution by monitoring walking abnormalities. Weapplied Principal Component Analysis into gait data to detect user's abnormalities that may indicate the progression of Parkinson's Disease. We validated our approach with a public database of foot sensor data, which includes vertical ground reaction force records of subjects with healthy gait and Parkinson's Disease patients. We used the euclidean distance asdata classifier. We reached a classification accuracy of 81.00% with leave-one-out cross-validation, which demonstrates the feasibility of our approach for tracking PD's symptoms based on user gait. All relevant data to reproduce our results are available in a public web page.

Parkinson's disease (PD) is a degenerative neurological disorder. It causes motor symptoms such as resting tremor, bradykinesia and gait disorders. The disease's progressive nature requires continuous monitoring of the motor symptoms to assist the neurologist in managing medication. With this purpose, Health Monitoring Systems (HMS) are used as a decentralized healthcare approach. On the other hand, most patients reject the current HMS solutions because they are invasive and stigmatizing. In this work, we present a non-invasive HMS for PD motor symptoms based on games. Because of the nature of games, the approach is able to collect data from patients without reminding them that they are under a disease's treatment. We validated our approach with 30 research subjects divided between PD group and Control group. We used Support Vector Machine (SVM) to identify the occurrence of PD's bradykinesia motor symptoms and reached a classification precision of 92.31%. Furthermore, 90,00% of the patients approved our HMS considering it as non-invasive and easily integrated into their routine.

One of the key challenges in constructing a Bayesian network BN is defining the node probability tables (NPT).For large-scale BN, learning NPT through domain experts knowledge elicitation is unfeasible.Previous works proposed solutions to this problem using the concept of ranked nodes; however, they have limited modeling capabilities or rely on BN experts to apply them, reducing their applicability.In this paper, we present an expert system based on production rules to define NPTs with the purpose of enabling the definition of NPTs by experts with no ranked nodes-specific knowledge.To create the rules, we elicited data from an expert in ranked nodes.To validate our approach, we executed an experiment with a BN already published in the literature to verify if, with our approach, a practitioner can achieve the same or better configuration for the NPTs.We used the Brier score to assess the NPTs accuracy and evaluated the results with the Wilcoxon test.All the Wilcoxon tests executed rejected the null hypotheses that stated that the Brier scores for the original NPTs method were the same as the new NPTs.By using our solution, a practitioner can accurately define NPTs without understanding the concept of ranked nodes.

Managing software development projects is a complex task because it requires organizing and monitoring several activities.Recently, in order to assist on software projects management, artifact-based models were proposed in the literature.However, the current solutions do not present means to monitor projects health and assist on decisions making.Due to the recent popularization of agile methods, they are the units of study of this research.In this work, we present a method to build artifact and measurementbased models to assess agile projects health.We applied the method to build a generic model based on industrys best practice.We defined the models artifacts and metrics based on findings of a literature review and the assistance of an expert.For each models artifact, we applied the Goal-Question-Metric paradigm to define the metrics.Afterwards, from the GQM meta-model, we constructed a Bayesian network.We validated the model with simulated scenarios.Given the successful results, we concluded that the method and model are promising.