01.1 - Graduação (Sede)
URI permanente desta comunidadehttps://arandu.ufrpe.br/handle/123456789/2
Navegar
12 resultados
Resultados da Pesquisa
Item Comparison of recommendation algorithms for user groups: a food-based case study(2023-04-24) Vasconcelos, Caio Giovanni Pereira; Silva, Douglas Véras e; http://lattes.cnpq.br/2969243668455081; http://lattes.cnpq.br/4775036700843482Item Utilização de filtragem colaborativa no auxílio de recomendação personalizada para leitores de mangá(2024-03-04) Brochardt, Rodrigo Nativo do Brasil; Garrozi, Cícero; http://lattes.cnpq.br/0488054917286587This study investigated, developed, and compared two approaches for generating manga recommendations: the Singular Value Decomposition (SVD) model and the Pearson Correlation Coefficient. The methodology involved data preparation through the development and execution of a web scraper to extract manga information and reviews from a highly active internet forum. Challenges arising in the applicability of these data extraction methods were addressed, along with alternatives for handling source blocking situations, model training, and performance evaluation, focusing on collaborative filtering and personalized recommendations for user profiles and manga works. In the implementation of SVD, latent patterns in user review data were identified, enabling personalized recommendations based on individual preferences through the sharing of experiences with similar profiles. However, metrics such as Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) revealed the need for model refinement to improve its accuracy, as well as alternative implementations for conducting comparisons and metrics related to the specific data mass used in the study. Conversely, the approach based on the Pearson Correlation Coefficient prioritized similarity between manga reviews to generate item-focused recommendations, although it significantly relied on the number of available reviews. This methodology offered a direct and valid logic for personalized recommendations based on relationships derived from reviews. The conclusion highlighted the future possibility of exploring hybrid methods combining the advantages of SVD and the Pearson Correlation Coefficient to achieve more precise and comprehensive recommendations, as well as validating techniques that bring different recommendation approaches for tangible comparison. The utilization of additional data gathered in the generated data mass to enrich the quality of recommendations was suggested, aiming to use more detailed parameters in recommendations, along with the employment of indirect approaches, such as using LLMs to aid in the recommendation process. Finally, the study emphasizes the importance of advancing these recommendation technologies to facilitate readers' lives by assisting in filtering the vast content offered by the industry and the internet.Item Análise de um sistema de recomendação de restaurantes sensível ao contexto sobre o grau de satisfação dos usuários(2023-09-01) Melo Filho, Carlos Olimpio Rodrigues de; Silva, Douglas Véras e; http://lattes.cnpq.br/2969243668455081; http://lattes.cnpq.br/6986499479035317Popular applications of recommender systems can be found in many areas. In the food business, platforms such as TripAdvisor stand out for suggesting specialized restaurant recommendations based on various types of relevant information, such as reviews from other users for the menu, atmosphere and recommendations for the closest restaurants are some of the specialties of these platforms. With the possibility of using new data sensitive to the user’s context, the main objective of this work is to evaluate the usage of the reason of going to the restaurant to reorganize the final restaurants recommendation through a context-based post-filtering. To achieve the goal, a mobile application was developed, the SR Recife Restaurants, to assess the degree of satisfaction of real users to the recommended restaurants, an online evaluation approach, using questionnaires, was used. When carrying out the experiment with 15 users, it was possible to notice an increase of 26.67% in the degree of satisfaction of the top-5 first recommendations when using the trip type to the restaurant as context data for the post-filtering phase.Item Sistema de recomendação de restaurantes, baseado em três tipos diferentes de filtragem de dados, nos bairros do Recife-PE(2021-12-20) Silveira Filho, Daniel Guilherme da; Cysneiros Filho, Gilberto Amado de Azevedo; http://lattes.cnpq.br/0534822491953359Considering that the city of Recife is in third place for gastronomic centers in Brazil, and that currently the amount of information present on the internet is bordering on infinity, tourists and even the local population go to the most popular restaurants. A consequence of this is that small accessories, which do not have a strong publicity for their brand, end up going out of business in the first five years of existence. Therefore, a research was carried out on recommendation systems, an analysis purpose for the development of a restaurant recommendation system, based on three different types of data filtering techniques, in the neighborhoods of the city of Recife. Therefore, it is necessary to explain the fundamentals of the filtering techniques that will be used in the development of the system, in addition to identifying how to define, collect, treat and analyze the data provided for the construction of the system, analyze the development of a recommendation-based system in the content, analyze the development of a recommendation system, based on the user, analyze the development of a hybrid recommendation system and finally identify which recommendation technique best results. A study is then carried out on the database, selection, collection and processing of data provided for the construction of the system, in addition to modeling a sample of people and results to be recommended, a study was also carried out on recommendation systems, definition, emergence and main filtering techniques, used in similar works, thus it was necessary to define which attributes and parameters to be collected, in addition to a modeling of the data capture form and development of the three selected collaborative filtering techniques, and finally, system tests are carried out to prove its functioning and analysis of collected data. Therefore, it appears that although the content-based filtering technique stands out in the results, the difference between this type of filtering and collaborative filtering was not significant, which imposes the observation that the recommendation system, based on neighborhoods , brings ease and convenience to users, in addition to promoting the brand of neighborhood restaurants and expanding tourism in the city of Recife, but it requires more interaction and information from registered users, to become more precise in their recommendations.Item Inferência automática de nível de dificuldade de receitas culinárias usando técnicas de processamento de linguagem natural(2020-12-21) Britto, Larissa Feliciana da Silva; Pacífico, Luciano Demétrio Santos; Ludermir, Teresa Bernarda; http://lattes.cnpq.br/6321179168854922; http://lattes.cnpq.br/9521600706234665; http://lattes.cnpq.br/5058497100007411In this work, a tool for inferring the degree of difficulty of cooking recipes will be proposed. The inference will be made by the textual classification of the recipe preparation methods. The tool will be a fundamental piece to the development of a contextaware contentbased cooking recipe recommendation system. Some of the main classifiers in Text Classification literature will be adopted, in addition to different feature extraction methods. An experimental evaluation is performed, in order to select the best approaches to compose the system.Item Recomendação e geração de receitas baseada na substituição de ingredientes(2020-12-21) Oliveira, Emília Galdino de; Pacífico, Luciano Demétrio Santos; Ludermir, Teresa Bernarda; http://lattes.cnpq.br/6321179168854922; http://lattes.cnpq.br/9521600706234665; http://lattes.cnpq.br/6278486720525640Nowadays, even with the increasing number of recipe sharing websites and systems, users may have difficulty to search specific dishes through the massive amount of data contained in such repositories. Also, finding recipes which best fit a handy set of ingredients, while at the same time contemplate some user wishes and restrictions, may become a very time consuming or even impossible task. In this work, we propose a new recipe recommendation and generation system, based on the substitution of recipe ingredients and a datadriven approach, in an attempt to help users finding a recipe that contemplates both their desires and food restrictions, avoiding food wastes.Item Avaliação entre algoritmos de filtragem colaborativa baseada em vizinhança e transferência de conhecimento para CD-CARS(2019) Silva, Guilherme Melo da; Silva, Douglas Véras e; http://lattes.cnpq.br/2969243668455081; http://lattes.cnpq.br/7122596102314881Recommendations in scenarios with the lack of preferences expressed by users is an importantlimitation for Recommendation Systems (RS). Due to this problem, cross-domain RS (CDRS)searches have gained relevance, where collaborative filtering (CF) is one of the most exploitedtechniques in this area. The CD-CARS system shows that the use of contextual information,available in user preferences, can optimize CF neighborhood-based algorithms, a techniquewidely used in multidomain CF. Although they provide accurate recommendations, some neigh-borhood-based algorithms such as the one used in the CD-CARS have the limitation of the useof multi-domains only in the occurrence of user overlap between domains, a non-trivial scenarioin real databases. This work presents a comparative analysis of different recommendation algo-rithms involving collaborative filtering techniques. The CD-CARS’ NNUserNgbr-transClosure(CF neighborhood-based) and Tracer (CF transfer learning-based) algorithms, were used as thebasis for the recommendation algorithms. In the experiments, the CF algorithms were integratedinto the context-aware techniques, addressed in the CD-CARS: Contextual Pre-Filtering andPost-Filtering, being applied on two data sets, formed by two auxiliary domains and one target,with and without overlap between domains. The MAE and RMSE performance metrics wereused to evaluate the algorithms. The results of the experiments showed that the Tracer algorithmpresented better results concerning the NNUserNgbr-transClosure algorithm in all experimentscenarios without user overlap, with and without the use of the Contextual Pre-Filtering or Post-Filtering.Item Suporte à decisão multicritério em aplicativos de saúde sob demanda(2019) Pereira, Gustavo Magalhães; Albuquerque Júnior, Gabriel Alves de; Falcão, Taciana Pontual da Rocha; http://lattes.cnpq.br/5706959249737319; http://lattes.cnpq.br/1399502815770584; http://lattes.cnpq.br/6456769669695121Health on demand applications have the main purpose of finding a doctor and take him to your home to provide home care for those who have limited mobility and seek a more convenient medical service, who do not want to face waiting lines and who wish to avoid to go to a hospital to treat basic illnesses. The technological advance has transformed the way traditional services are offered on demand, which is increasingly popular in Brazil. The Federal Council of Medicine (CRM), knowing the impact of technological advances in the practice of medicine, published a resolution No. 2178/2017, which seeks to regulate the operation of applications that offer medical consultation at home. According to the resolution, all applications that offer this service are required to provide a list of physicians available to the patient to choose the best doctor to take care of their case, but the applications do not offer patient assistance in the decision and is in need of a computational solution. In this work was carried out the planning and development of a recommendation system using the methodology to support multi-criteria decision analysis. It was used as a case study the application Clinio, a product of health on demand developed by Epitrack. The solution applied to recommend the best physicians has the purpose of helping the users of the application in choosing the professional that best fits their needs and preferences. To do so, it was used recommendation algorithms to select doctors based on symptoms and geolocation and the Analytical Hierarchical Process (AHP), based on criteria to classify them such as the value of the consultation, the distance between the doctor and the patient, and the age of the physician. The system was implemented using a database of 143 doctors from Pernambuco who work in 10 clinical cases. Through the tests performed it was observed in the recommendation obtained by the users that the system assists in the process of choosing the best professional for a case through the preferences definitions.Item Um sistema de recomendações de eventos culturais com áudio-descrição(2018) Souza Filho, Robson Ugo Ferreira; Medeiros, Victor Wanderley Costa de; Falcão, Taciana Pontual da Rocha; http://lattes.cnpq.br/5706959249737319; http://lattes.cnpq.br/7159595141911505Around one billion people in the world live with some kind of disability, while almost 24% of the Brazilian population has declared some kind of disability in the 2010 CENSUS. Therefore, the increasingly constant presence of people with visual impairment in cutural spaces has increased and is also due to the presence of communicational accessibility resources. The technological advance has made the disposition of such resources much easier and closer, increasing audience’s autonomy. Based on this argument, this work aims to present the planning and development of an audio-description cultural event recommendation system for blind people using mobile devices, increasing the independence and capacity of the natural indication process to human relationships through collaborative filtering item-based and content-based algorithms. We generated a potential database of these kinds of events, a study about the proposed algorithms and an application usability experiment. The absence of statistical evaluation for the maden evaluations was discovered, since the users’ approach in selecting notes for the events is personal, also resulting from the non-existent consistency of its data.Item Um estudo comparativo de técnicas para a classificação contextual de companhia para sistemas de recomendação sensíveis a contexto(2019-01-22) Silva, Douglas Henrique Santana da; Silva, Douglas Véras e; http://lattes.cnpq.br/2969243668455081; http://lattes.cnpq.br/6428879549861854Nowadays, the vast amount of information has harmed users during decision making. In face of this problem, recommendation systems have been proposed in order to offer suggestions that help users to overcome such problem. These suggestions are even more valuable when these systems begin to suggest items based on the user contexts. Among these contexts, the companion context can be highlighted. Through the inference of the companion context the system may suggest different items if the user is accompanied or not. An example of a system that has such features is the CD-CARS. However, the unsupervised learning method for companion inference on CD-CARS has some limitations. In this way, the present research analyzed and highlighted a supervised learning method that can replace the current company contextual classification approach executed in the CD-CARS.