Browsing by Author "Cunha, Carlos"
Now showing 1 - 10 of 17
Results Per Page
Sort Options
- Agile-based Requirements Engineering for Machine Learning: A Case Study on Personalized NutritionPublication . Cunha, Carlos; Oliveira, Rafael; Duarte, RuiRequirements engineering is crucial in developing machine learning systems, as it establishes the foundation for successful project execution. Nevertheless, incorporating requirements engineering approaches from traditional software engineering into machine learning projects presents new challenges. These challenges arise from replacing the software logic derived from static software specifications with dynamic software logic derived from data. This paper presents a case study exploring an agile requirement engineering approach popular in traditional software projects to specify requirements in machine learning software. These requirements allow reasoning about the correctness of software and design tests for validation. The absence of software specification in machine learning software is offset by employing data quality metrics, which are assessed using cutting-edge methods for model interpretability. A case study on personalized nutrition and physical activity demonstrated the adequacy of user stories and acceptance criteria format, popular in agile projects, for specifying requirements in the machine learning domain.
- Are you lost? Using Facial Recognition to Detect Customer Emotions in Retail StoresPublication . Borges, Valter; P. Duarte, Rui; Cunha, Carlos; Mota, David; ThinkmindThe understanding of consumer behavior is a dynamic field, especially relevant to the success of companies and for consumer satisfaction. It is especially important in the situation of intense competition, currently characteristic for the retail store industry, where companies fight for every individual customer. Moreover, companies do not want customers to enter their system and leave without buying products they intended to buy. This has an impact on user satisfaction and retail stores income. In this paper we present a method that targets customer satisfaction in the aforementioned context using a facial recognition system acting at the emotional level of the customer. Our method is based on cumulative negative emotions that are associated to a sadness level, which triggers events for retail store assistants to help customers. Results show that this method is adequate to measure these emotions and is a useful reference for retail store assistant intervention.
- Automated Reusable Tests for Mitigating Secure Pattern Interpretation ErrorsPublication . Cunha, Carlos; Pombo, NunoThe importance of software security has increased along with the number and severity of incidents in recent years. Security is a multidisciplinary aspect of the software development lifecycle, operation, and user utilization. Being a complex and specialized area of software engineering, it is often sidestepped in software development methodologies and processes. We address software security at the design level by adopting design patterns that encapsulate reusable solutions for recurring security problems. Design patterns can help development teams implement the best-proven solutions for a specialized problem domain. However, from the analysis of three secure pattern implementations by 70 junior programmers, we detected several structural errors resulting from their interpretation. We propose reusable unit testing test cases based on annotations to avoid secure pattern interpretation errors and provide an example for one popular secure pattern. Providing these test cases to the same group of programmers, they implemented the pattern without errors. The reason is annotations build a framework that disciplines programmers to incorporate secure patterns in their applications and ensure automatic testing.
- Automatic Camera Calibration Using a Single Image to extract Intrinsic and Extrinsic ParametersPublication . P. Duarte, Rui; Cunha, Carlos; Pereira Cardoso, José CarlosThis article presents a methodology for accurately locating vanishing points in undistorted images, enabling the determination of a camera's intrinsic and extrinsic parameters as well as facilitating measurements within the image. Additionally, the development of a vanishing point filtering algorithm is introduced. The algorithm's effectiveness is validated by extracting real-world coordinates using only three points and their corresponding distances. Finally, the obtained vanishing points are compared with extrinsic parameters derived from multiple objects and with intrinsic parameters obtained from various shapes and images sourced from different test sites. Results show that through a single image, the intrinsic parameters are extracted accurately. Moreover, Using 3 points to determine the extrinsic parameters is an excellent alternative to the checkerboard, making the method more practical since it does not imply the manual positioning of the checkerboard to perform the camera calibration.
- Automatic User Testing and Emotion Detection in Interactive Urban DevicesPublication . P. Duarte, Rui; Cunha, Carlos; Pereira Cardoso, José CarlosAutomated testing and evaluation of interfaces is a well-established reality supported by many tools that shorten the time to deploy new software versions to the user. However, exploring users’ emotions while interacting with interfaces as a tool to further increase the quality of traditional usability evaluation methods is still far from being a reality. This work uses the automatic analysis of users’ emotions while interacting with touchable interactive urban devices to detect usability issues. To this end, a coupled approach is implemented: the data is acquired from the interaction, and user emotions are extracted and processed to determine the emotional status during the interaction. This data is integrated into a web application so that designers can further improve the quality of the interface in the presence of negative emotions. Results show that the experimental tests showed that different users manifest similar negative emotions in the same contexts, which is a clear sign of usability issues that are to be corrected by the design team.
- Behavioral Anomaly Detection of Older People Living IndependentlyPublication . Cunha, Carlos; P. Duarte, Rui; Mota, DavidOlder people living independently represent one significant part of the population nowadays. Most of them have family or friends interested in being informed about changes in their routine. Considering these changes signal some physical or mental problem, they can trigger a contact or action from the interested persons to provide support. This paper presents an approach for non-intrusive monitoring of older people to send alerts after detecting anomalous behaviors. An analysis of seven months of data gathered using PIR sensors in a couple’s living house has shown regularities in their presence in compartments along the day. We validated the adequacy of an outlier detection algorithm to build a model of the persons’ behavior, exhibiting just 3.6% of outliers interpreted as false positives.
- Enhancing quality of life: Human-centered design of mobile and smartwatch applications for assisted ambient livingPublication . Augusto, Gonçalo F.; P. Duarte, Rui; Cunha, CarlosBackground: Assisted ambient living interfaces are technologies designed to improve the quality of life for people who require assistance with daily activities. They are crucial for individuals to maintain their independence for as long as possible. To this end, these interfaces have to be user-friendly, intuitive, and accessible, even for those who are not techsavvy. Research in recent years indicates that people find it uncomfortable to wear invasive or large intrusive devices to monitor health status, and poor user interface design implies a lack of user engagement. Methods: This paper presents the design and implementation of non-intrusive mobile and smartwatch applications for detecting older adults when executing their routines. The solution uses an intuitive mobile application to set up beacons and incorporates biometric data acquired from the smartwatch to measure bio-signals correlated to the user’s location. User testing and interface evaluation are carried out using the User Experience Questionnaire (UEQ). Results: Six older adults participated in the evaluation of the interfaces. Results show that users found the interaction to be excellent in all the parameters of the UEQ in the evaluation of the mobile interface. For the smartwatch application, results vary from above average to excellent. Conclusions: The applications are intuitive and easy to use, and data obtained from integrating systems is essential to link information and provide feedback to the user.
- Knowledge retention through observation of instant messaging systemsPublication . Costa, João; P. Duarte, Rui; Cunha, Carlos; Henriques, JoãoKnowledge is the most valuable asset in today’s organizations. Since it offers an unbeatable competitive advantage, valuable knowledge demands strict management principles to avoid being lost. Instant messengers provide an opportunity to gather knowledge passed through individuals in the organization. By modeling that knowledge using machine learning techniques, it becomes possible to retain and make it ubiquitous throughout the organization. This paper presents a solution for gathering, modeling, and retrieving knowledge associated with the technical support in organizations, using machine learning algorithms. The solution comprises the architecture, data preparation techniques and machine learning algorithms. The experimental evaluation exhibits the algorithms with better performance for this class of problems.
- Machine Learning and Food Security: Insights for Agricultural Spatial Planning in the Context of Agriculture 4.0Publication . Martinho, Vítor; Cunha, Carlos; Pato, Lúcia; Costa, Paulo Jorge; Sánchez-Carreira, María Carmen; Georgantzís, Nikolaos; Rodrigues, Raimundo Nonato; Coronado, Freddy: Climate change and global warming interconnected with the new contexts created by the COVID-19 pandemic and the Russia-Ukraine conflict have brought serious challenges to national and international organizations, especially in terms of food security and agricultural planning. These circumstances are of particular concern due to the impacts on food chains and the resulting disruptions in supply and price changes. The digital agricultural transition in Era 4.0 can play a decisive role in dealing with these new agendas, where drones and sensors, big data, the internet of things and machine learning all have their inputs. In this context, the main objective of this study is to highlight insights from the literature on the relationships between machine learning and food security and their contributions to agricultural planning in the context of Agriculture 4.0. For this, a systematic review was carried out based on information from text and bibliographic data. The proposed objectives and methodologies represent an innovative approach, namely, the consideration of bibliometric evaluation as a support for a focused literature review related to the topics addressed here. The results of this research show the importance of the digital transition in agriculture to support better policy and planning design and address imbalances in food chains and agricultural markets. New technologies in Era 4.0 and their application through Climate-Smart Agriculture approaches are crucial for sustainable businesses (economically, socially and environmentally) and the food supply. Furthermore, for the interrelationships between machine learning and food security, the literature highlights the relevance of platforms and methods, such as, for example, Google Earth Engine and Random Forest. These and other approaches have been considered to predict crop yield (wheat, barley, rice, maize and soybean), abiotic stress, field biomass and crop mapping with high accuracy (R2 ≈ 0.99 and RMSE ≈ 1%)
- Meal Suggestions for Caregivers and Indecisive Individuals Without a Set Food PlanPublication . Cunha, Carlos; Cardoso, Tiago R.; P. Duarte, RuiRecommendation systems have played a crucial role in assisting users with decision-making across various domains. In nutrition, these systems can provide valuable assistance by offering alternatives to inflexible food plans that often result in abandonment due to personal food preferences or the temporary unavailability of certain ingredients. Moreover, they can aid caregivers in selecting the most suitable food options for dependent individuals based on their specific daily goals. In this article, we develop a data-driven model using a multilayer perceptron (MLP) network to assist individuals in making informed meal choices that align with their preferences and daily goals. Our study focuses on predicting complete meals rather than solely on predicting individual food items since food choices are often influenced by specific combinations of ingredients that work harmoniously together. Based on our evaluation of a comprehensive dataset, the results of our study demonstrate that the model achieves a prediction accuracy of over 60% for an individual complete meal.