Titre : | Community Detection in Social Media (Twitter | Type de document : | projet fin études | Auteurs : | Lamalem Nizar, Auteur | Langues : | Anglais (eng) | Catégories : | Internet des Objets et Services Mobiles ( IOSM )
| Mots-clĂ©s : | Community Detection, Hortonworks, Oozie, Flume, Hive, Louvain, Social Graph. | Index. dĂ©cimale : | mast 67/18 | RĂ©sumĂ© : | The aim of this internship is to create a big data application for Community Detection in Twitter Social network based on user’s keywords and features selection. The target was to create the biggest possible Social Graph to get the best result. We started first by deploying a cluster of 3 nodes to gather the data using Hortonworks tool Flume. We refined the gathered data and load it into hive, and from hive to the graph database (Neo4j). All of this in 24/7 cyclic continuous way using Oozie as a workflow Coordinator. For each user search, we apply the Louvain algorithm to detect communities on the subgraph that suit the search parameters and display the result in a web application. This report presents the different steps done to implement this project
|
Community Detection in Social Media (Twitter [projet fin études] / Lamalem Nizar, Auteur . - [s.d.]. Langues : Anglais ( eng) Catégories : | Internet des Objets et Services Mobiles ( IOSM )
| Mots-clĂ©s : | Community Detection, Hortonworks, Oozie, Flume, Hive, Louvain, Social Graph. | Index. dĂ©cimale : | mast 67/18 | RĂ©sumĂ© : | The aim of this internship is to create a big data application for Community Detection in Twitter Social network based on user’s keywords and features selection. The target was to create the biggest possible Social Graph to get the best result. We started first by deploying a cluster of 3 nodes to gather the data using Hortonworks tool Flume. We refined the gathered data and load it into hive, and from hive to the graph database (Neo4j). All of this in 24/7 cyclic continuous way using Oozie as a workflow Coordinator. For each user search, we apply the Louvain algorithm to detect communities on the subgraph that suit the search parameters and display the result in a web application. This report presents the different steps done to implement this project
|
|