DISIT Twitter Vigilance

SmartCityBigData2015: Overview Twitter Vigilance
SmartCityBigData2015, disit lab, 13 novembre 2015, Firenze


Twitter Vigilance, TV is a multi-user tool for Twitter analysis
Twitter Vigilance, TV is a multi-user tool for Twitter analysis, developed by the DISIT Lab of the University of Florence. According to our view, a Twitter Channel consists of a set of simple and complex queries performed on Twitter platform. The simplest Twitter Channel monitored can control tweets referred to a single user, to a single hashtag, or to a single keyword. Complex Channels consist of tens of complex queries according to Twitter syntax combined with keywords, users, hashtags, etc. The single query is not only a tool to search, but also a key to analysis. The Twitter Vigilance ensures the collection of 98% of tweets/retweets referred to events by providing the yield and precision evidence to the user. Twitter Vigilance is capable to monitor and analyse slow and explosive events on Twitter with same efficiency and precision. A fast or explosive event occurs with several hundred thousands of tweets per day/per hour. Slow events can occur with very few tweets per day or week or their absence. Twitter Vigilance provides adaptive algorithms to allow effectively cope with slow events that become explosive without losses, acquiring all tweets and retweets. The Twitter Vigilance tool is active since April 2015. Active Channels on Twitter vigilance as: EXPO 2015, CNR EXPO2015, Firenze, ApreToscana, ConsumoSuolo, weather, pharmacovigilance, laudatesi, maturity ECLAP and Europeana, Tuscany weather alert, civil protection, etc. A part of these analyses are available and tracked on public examples on http://www.disit.org/tv. Other Twitter Channels are reserved by researchers and analysts or for entities that are working on Twitter Vigilance according to contract agreement signed with the DISIT Lab. These operators can set their channels and queries, associate labels analysis trends for marking events and situations, can download data, get special draws, print charts in various formats.


DISIT Twitter Vigilance is a multi user tool for making analysis of "Twitter channels". Each channel can be tuned to monitor one or more Search Queries on Twitter with a sophisticated and expressive syntax. The simplest query can be the single keyword. 
Presently, DISIT Twitter Vigilance is used in: (i) a join collaboration with LAMMA and CNR IBIMET for monitoring weather status and forecast related twitters for different purposes, (ii) several smart city projects (as Sii-Mobility SCN, RESOLUTE H2020, ..) for monitoring city services, critical events and conditions, and user behavior and opinion, (iii) city response to events, (iv) pharmaceutical drug response, (v) advertising impact, (vi) impact of tecnologies and information, (vii) response of population to stimuls, (viii) political appreciation, etc.
Channels under monitoring are presently a relevant number, while only some of them can be mentioned as: Aereoporto di Firenze, EXPO 2015, some advertising, allerta meteo toscana, caldo, ConsumoSuolo, paoloNesi, protezione civile toscana, apretoscana and matchmaking, ECLAP, laudatesi, drugs, vargroup, university of florence, etc. Others are private analysis of the reseachers that have organized the search on this tool.

Twitter Vigilance tool of DISIT Lab is capable to monitor and analyze with the same effectiveness and precision fast and slow events on twitters. Twitter vigilange guarantees the collection of 98% of tweets related to a given event even if it slow or very fast, explosive. Fast means events that provoke several hundreds thousands of twieets per day. Slow may means one or none per day or week. They are processed for sentiment (not publicly esposed), and a pletora of detailed analyses and reports are provided.

Thus, the channels under monitoring are accessible for public perusal (according to the decision of the channel curator) with limited capabilities via http://www.disit.org/tv/ and many other tools not accessible to public.

Register users of DISIT Twitter Vigilance tools and services may:
  • create one or more monitoring channels (canale), as reported in the figures; each channel can be tuned to monitor one or more Search Queries on Twitter with simple and sophisticated and expressive syntax. The simplest query can be the single keyword;
  • perform multiple channels, that may use the same Search Queries;
  • download collected data sets (under request);
  • decide to provide public access to their channels analysis (as in the channels accessible without registration) or NOT;
  • access at the history of tweets per channel, per search, per users, etc.;
  • perform full search analysis on the whole set of tweets with faceted capabilities on several aspects of the collected data;
  • access to statistic regarding precision by which tweets and RT are collected and stored;
  • print graphs and export them in figures in different formats;
  • perform analyses at level of channel, search, users, tweets, retweets, etc.:
    • trends of the Search Queries as reported in the above figure amd public;
    • distributions on population and activities of users (private);
    • distributions about other tags/keywords (private);
    • geographic distriobution of twitters of single or multiple channels;
    • distributions regarding tweet and re-tweets;
    • natural language processing (pratially accessible at public only on few demonstrative and public channels, such as Expo2015 and Nasa Pluto). Distribution of tags, citations, keywords, verbs, etc.
    • etc.

Full indexing of collected twitters via SOLR (ask to Paolo Nesi for access)

For additional information, or if you are interested to set up and curate your own channels for monitoring Twitter. Stay tuned, since a number of new and exciting features are in progress and will be accessible soon.
For any information and agreement, please send an email to DISIT lab coordinator: Paolo.nesi@unifi.it

last update: 04-September-2015