| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 4.63 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Este projeto está inserido num ramo de inteligência artificial chamado Machine Learning e este baseia-se na ideia de que os sistemas possam aprender com informação, identificar padrões e por sua vez tomar decisões com o mínimo de intervenção humana. Machine Learning é utilizado no dia-a-dia em recomendações online de produtos, deteção de fraudes, anúncios em tempo real, reconhecimento de voz e texto, entre outros. A presente dissertação tem como objetivo documentar todo o processo de implementação de um sistema de recomendação de anúncios em tempo real na área do desporto. O sistema de recomendação (baseado em conteúdo) com base no perfil de cada utilizador associa os anúncios desportivos que correspondem com a sua procura ou oferta e envia-lhe uma notificação. Os utilizadores têm acesso às características dos anúncios, mas só poderão ver o proprietário do anúncio e entrar em contacto com ele se usufruírem de uma conta premium. Este sistema permite aos utilizadores criarem e visualizarem anúncios desportivos em várias modalidades, assim estes poderão analisar as melhores ofertas ou procuras atualmente no mercado. O sistema de recomendação é composto por uma solução web desenvolvida em ASP.NET MVC e uma solução móvel desenvolvida em React Native e visa promover uma nova abordagem do processo de captação de jogadores e treinadores.
This project is embedded in an artificial intelligence branch called Machine Learning and this is based on the idea that systems can learn from information, identify patterns and make own decisions with minimal human intervention. Machine Learning is used daily in online product recommendations, fraud detection, realtime announcements, voice and text recognition, among others. The present dissertation aims to document the entire process of implementing a realtime announcements recommendation system in sport area. The recommendation system (content-based) based on the profile of each user associates the sports announcements that match with his search or offer and sends him a notification. Users have access to the features of the announcements but will only be able to see the owner of the announcement and get in touch with him if they have a premium account. This system allows users to create and view sports announcements of different modalities, so they can analyze the best deals or searches currently on the market. The recommendation system consists of a web solution developed in ASP.NET MVC and a mobile solution developed in React Native and aims to promote a new approach to the process of capturing players and coaches.
This project is embedded in an artificial intelligence branch called Machine Learning and this is based on the idea that systems can learn from information, identify patterns and make own decisions with minimal human intervention. Machine Learning is used daily in online product recommendations, fraud detection, realtime announcements, voice and text recognition, among others. The present dissertation aims to document the entire process of implementing a realtime announcements recommendation system in sport area. The recommendation system (content-based) based on the profile of each user associates the sports announcements that match with his search or offer and sends him a notification. Users have access to the features of the announcements but will only be able to see the owner of the announcement and get in touch with him if they have a premium account. This system allows users to create and view sports announcements of different modalities, so they can analyze the best deals or searches currently on the market. The recommendation system consists of a web solution developed in ASP.NET MVC and a mobile solution developed in React Native and aims to promote a new approach to the process of capturing players and coaches.
Description
Keywords
Sistema de recomendação Desporto ASP.Net MVC React Native SignalR FireBase Recomendation system Sport ASP.Net MVC React Native SignalR FireBase