In the last decade, a large amount of data has been published in different fields\nand can be used as a data source for research and study. However, identifying\na specific type of data requires processing, which involves machine learning\nclassifying techniques. To facilitate this, we propose a general framework that\ncan be applied to any social media content to develop an intelligent system.\nThe framework consists of three main parts: an interface, classifier and analyzer.\nThe analyzer uses media recognition to identify specific features. Then,\nthe classifier uses these features and involves them in the classification\nprocess. The interface organizes the interaction between the system components.\nWe tested the framework and developed a system to be applied to image-\nbased social media networks (Instagram). The system was implemented\nas a mobile application (My Interests ) that works as a recommendation and\nfiltering system for Instagram users and reduces the time they spend on irrelevant\ninformation. It analyzes the images, categorizes them, identifies the interesting\nones, and finally, reports the results. We used the Cloud Vision API\nas a tool to analyze the images and extract their features. Furthermore, we\nadapted support vector machine (SVM), a machine learning method, to\nclassify images and to predict the preferred ones.
Loading....