Social Media
The IntuView technology may be applied to the social media - Tweets, Facebook etc in a manner that provides an opportunity for both the defense and the commercial areas.
IntuScan processes social media messages. This is a significant value since automated systems of entity and sentiment recognition that are trained on standard language texts produce poor results on such media. The IntuView Social Media Monitor is an advanced tool for measuring issues and sentiment in social media. The application of the IntuView Natural Language Processing technology provides a comprehensive picture of the topics of discourse, sentiment towards persons and z, breakdown of ideological and political sympathies, relations, and more.
IntuScan can cluster social media associated with subgroups based on their linguistic and interest affinities to facilitate further collection efforts relating to areas of focus of the different groups and to analyze the sentiment of the group members towards different entities, products and issues, as well as to drill down to sentiment towards specific attributes or features of the entities and aggregated sentiment towards parent entities of a number of entities.
This enables the user to identify: (1) “Who is talking” (sociologically, ethnically, based on their social frames of reference); (2) “How many” (the scope of the situation); (3) “What and whom are they are talking about” (entity recognition, disambiguation and resolution, including “idea entities”); (4) “Where they are” - not only according to meta-data but by content; (5) “How and why do they feel about the object” - sentiment analysis.
The IntuView Social Media Monitor can be applied to mapping of individuals, organizations, and places that are in the focus of the various groups and the elements that are involved in the discourse and the co-relationships between them. These relationship maps provide not only a static picture but also a basis for analysis over a period of time. These can facilitate political research during elections and ongoing analytics of groups within a country (relations of command and control, hostility, family-tribe, mentors-follower, etc.)
Uses of IntuScan for social media include:
-
Mining public opinion relating to products in order to improve marketing.
-
Forging political campaign messages through identification of defined sentiment towards issues and persons in specific sectors on the social web.
-
Targeting commercial messages to well defined social sectors based on their implicit (not declared) sympathies.
-
Identifying social trends in defined sectors in order to exploit them in commercial motifs.
-
Identifying social dissatisfaction in order to predict and measure social unrest and potential political turmoil.
-
Extracting intelligence for HLS and CT.
-
Efficient networking by characterization of forums and persons in them in order to direct clients to the areas in the social cyberspace where they can connect with useful contacts.
-
HLS crisis management - efficient extraction of information from social media can facilitate rescue and stabilization operations in the crisis.
-
Verification of candidates for work.
-
Lead Generation - identification in Social Media of sectors and entities that may be compatible with the client's product.
-
Strategic intelligence on unrest in foreign countries (Iranian "Green Revolution", "Arab Spring").
-
Mapping of traffic on the social web for identification of ideas that occur in different social and linguistic contexts (migration of ideas in the context of revolutionary thought, radicalization etc.
The collection of the data may be implemented in the following ways:
-
The client may provide the data based on subscription to mainstream media or acquisition of feeds directly from Twitter/Facebook/Linkedin etc. through their official sources (GNIP and others). This option entails additional costs for acquisition of the firehose.
-
IntuView can set up crawlers for the capture of the streams of data based on the client’s requirements. This option will be integrated into a full proposal if requested.
The analysis includes:
-
Identification of prominence of entities (persons, groups, organizations, countries, events, and issues) in the media discourse in the relevant theatres (e.g. –what are the key topics, domains, issues, persons, etc. that feature in the discourse).
-
Identification of sentiment towards the above-mentioned entity types (positive/negative sentiment) including drill down to the key characteristics that the sentiment refers to (an entity’s intelligence, honesty, etc.)
-
Drill down to the levels of reference and the sentiment according to combinations of identified parameters: source, source country, source language, source author, source ideological-political characterization, time frame, etc.
-
User-enabled update of the knowledge base to reflect ongoing concerns and entities (e.g. new concepts or ideas that are floated, particular individuals of interest who appear on the political scene).