Directing Intelligence creates adaptive, collaborative intelligent platforms using machine learning to process enterprise data and uncover knowledge. Their Datactif® SoNetA platform combines social network analysis and text mining to help enterprises increase profitability through applications like market research, brand management, audience segmentation, and crisis management. It identifies communities, influencers, sentiments, concepts and clusters within social data to provide insights into customers, campaigns, and competitive intelligence.
2. Directing Intelligence
Knowledge is not only vital for the enterprise profitability but an integral
part of its core business process no matter the sector of its operability.
Business Intelligence as entity that processes and transforms information
into knowledge, must be in the centre of Business.
DIRECTING INTELLIGENCE, since 2000, bridges the gap between
business and intelligence by creating the foundations for an Intelligent,
Operational, Evolutionary Enterprise Ecosystem.
Aligned to each Enterprise Ecosystem, DIRECTING INTELLIGENCE
designs and creates adaptive, collaborative Intelligent Open Architecture
Platforms based on machine learning methodology and algorithms (neural
network, fuzzy systems, genetic algorithms, SVM, etc…), platforms that
process transactional data from operating systems and data from the
Web, numerical as well as unstructured data, allowing real time and on
line, substantive assessment of enterprise corporate knowledge.
4. Communities Detection
Social network analysis is the mapping and measuring of relationships and
flows among people, groups, organizations, computers, URLs, and other
connected information/knowledge entities.
The community value is incomparably more important than the score of any
individual customer, no matter how good it may be. And this is because
communities react randomly and chaotically and thus an event (purchase
of a product, subscription to a new service, positive comment, complaint or
churn) creates chain reactions with exponential effect.
Identify and understand communities help to test, predict and create
efficient and cost effective business and communication strategy.
Influencers identification is the number one objective in every social
media as specific users exercise influence over an organization and its
potential customers. Influencers are activists, well-connected, have
impact, have active minds, and are trendsetters, though this set of
attributes is aligned specifically to consumer markets.
Targeting influencers, is seen as a means of amplifying marketing
messages in order to counteract the growing tendency of prospective
customers to ignore traditional marketing efforts.
INFLUENCE ANALYSIS & RECOMMENDER SYSTEM for a famous
holidays destination. Visits to restaurants proposed by influencers through
published photos in FB & Instagram, have increased.
Influence Analysis
5. Sentiment Analysis
Sentiment Analysis retrieves user reviews and lists their comments in
Positive, Negative or Neutral Probability. But more important is the further
analysis and evaluation of reasons of positive, neutral and negative
indicators. Using Term analysis and Fact extraction we can identify and
understand the reasons (some times unknown to an enterprise) for clients’
satisfaction and dissatisfaction.
Promotional Campaigns evaluation through clients satisfaction
for a supermarket chain.
Terms analysis and Fact extraction identified reasons of negative
commentsfor a company. Focus on Influencers, we noticed that their
negative comments create another classification. They are unsatisfied
mostly by the company's’communication(they want a more personalized
one) and this feeling influences negatively their trust to the company
6. Concept Extraction
Concept extraction identifies the emergent semantics that characterize the
content of users communications. Concepts are extracted from available
comments by employing supervised learning algorithms and natural
language processing techniques.
Example of a Luxury Hotel. Focusing on positive comments
we extracted concepts such as “breathtaking view”, “helpful staff”,
“recommended hotel”, “clean and comfortable room” , etc…
Polarization is the process by which a social group is divided into two
opposing sub-groups that have conflicting and opposing positions, goals
and opinions. The understanding of the polarization phenomenon creation
(and what issues are the cause) is important because it is the key to
opinion understanding. Besides, a holder of a biased opinion is likely to
hold the same extreme view over time and in other issues.
A company published a POST with a promotional offer (free ticket
for a Zoological Park). The small but relatively compact group of animal
lovers rose up, accusing the company of insensitivity towards
the imprisoned animals
Polarization
7. Clustering
Based on sentiment analysis, authority score and users attitude in a given
forum, Clustering allows to discover groups (clusters) of users with
common characteristics. Social Media clusters are :
Superfans. Users with the highest activity and most positive attitude.
Should leverage their enthusiasm to spread the news and talk about new
products. They must always be kept informed and encouraged in their
positive attitude.
Critical Thinkers. Users with values of Authority and of Hub (with a
wider range of different users). They affect a sale even when they don’t
participate actively to a purchase process.
Neutral. These users need to be kept up to date in a non-intrusive way.
This means that they should not be acknowledged directly, but they might
be invited for example to special events for free. Neutral users are not
always neutral; they just balance out over time. So if we can get them to be
slightlymore positive that can only help the business.
Inactive. They are mostly interested in the product than the brand. But
there are a lot of users in this cluster that plan a future purchase, so they
must be considered as prospects.
8. Social Media Performance
Social media performance audit having as objective to consolidate a clear
view on ROI, uses social and digital data to report on performance for each
enterprise, its competition and market in general, across 5 important KPIs:
COMMUNITY
ENGAGEMENT
ADVOCACY
SHARE OF VOICE
REACH
Social Media Dashboards are designed ad hoc for each enterprise and can
be modified following business needs and competition environment.
DATACTIF® Suite of Big Data Analytics and DATACTIF® SoNetA,
create the bridge between transactional and Social Media data, offering
each enterprise state of science Real Big Data Analysis and through
PLANNER e-CRM, a single customer view.
DATACTIF Big Data