The main objective of the platform is to increase, at national level, the social integration of LGBTI people and to help Civil Organizations and Public Institutions to prevent homophobic and transphobic discrimination.

The ACCEPT platform operates by monitoring the sources at regular intervals, optimized on the basis of to the different types of opinion flows. To automatically identify the emerging topics (topics), state-of-the-art tools are used in the field of natural language processing (NLP).

In addition, the system allows:

1. data extraction for periodic summarized structured documents, through dedicated APIs, exportable graphs, 2. online access to analysis functions, 3. the classification of tweets as porters or not of hate messages.

Software components

The ACCEPT platform can be used through a web dashboard for the collection, processing and display of data concerning hate narratives towards the LGBTI community.

The choice of the main components are privileged the modular nature of the system, the use of international standards recognized for transmission, the availability of advanced tools for analysis and the ability to manage large amounts of data.

The platform allows the collection of data regarding a list of topics of interest from the following web channels:

1. Post Facebook, tweet, through the official API made available by the various platforms;

2. News from Google News and Bing News, made available by search engines via RSS feeds.

The various data collection processes are automatically executed at pre-established time intervals.

The Platform is accessible from a standard browser at http://accept.arcigay.it/, while the extraction of data through API for the creation of reports is available at http://accept.arcigay.it/api.

The Admin area, which allows you to fine-tune data extractions, is accessible only by accredited users with username and password credentials. In the same way, only accredited users can proceed with the classification of news and tweets or post facebook according to what is indicated in the annotation guide.