Download Customer Sentiment Analysis

Customer Sentiment Analysis
Key Questions
1) How to evaluate the new mechanisms that influence decisionmaking processes?
2) How to identify the users that are closest to a product and that
cause the most sensitive issues as influencers?
3) In which areas or countries are substantial differences emerging
in brand and product perception?
4) What actions should be taken as a result?
2
Scenario
• New technologies have enabled a greater ease of circulating
information, and their rapid take-up by companies as tools for
customer communication and relationship offers both major
opportunities and issues requiring extremely careful handling.
• Consumers are increasingly using the Internet as a place for
discussion and sharing experiences, a point of reference especially
when they are considering the purchase of complex goods and
services.
3
2.0 Effects vs Brand Reputation
Based on what you have read \ seen \ heard on a SOCIAL NETWORK \ COMMUNITY, blog or
forum recently, your general view on ... (company) is ...
Deteriorated
Improved
Barilla
4
41
55
Ferrero
4
42
54
Coop
4
Esselunga
43
6
11
Enel
10
39
8
45
39
44
44
15
44
42
9
Intesa San…
45
47
11
Mediaset
43
49
13
42
46
7
41
53
Alitalia
23
Telecom Italia
23
Trenitalia
Rai
50
17
Pirelli
Benetton
51
44
Poste Italiane
Unicredit
52
43
Fiat
Eni
4
Unchanged
42
35
44
33
26
40
33
34
47
32
27
Based on who has heard, seen, or read company information on a blog. Forum, social network ... (CAWI)
Sentiment Analysis: Definition
• Sentiment analysis or opinion mining refers to a broad
(definition challenged) area of natural language processing,
computational linguistics and text mining. Generally speaking, it
aims to determine the attitude of a speaker or a writer with
respect to some topic.
• The attitude may be their judgment or evaluation (see appraisal
theory), their affective state (that is to say, the emotional state of
the author when writing) or the intended emotional
communication (that is to say, the emotional effect the author
wishes to have on the reader).
5
Sentiment analysis: From Wikipedia, the free encyclopedia
Customer Sentiment Monitoring
The design and the success of Customer Sentiment analysis are primarily
configured in the definition of the following areas:
• Define Goals and
objectives expected by
the web tracking and
analysis
• Identify drivers
representing the brand
that will be monitored
in order to meet goals
• Identify the audience
you want to listen and
target
• In digital media:
•What is my audience on Social Media?
•What kind of profile do they have?
•Where are they?
Goals &
Objectives
6
Drivers
Domains
Sentiment Survey Main Phases
The survey of sentiment and evaluation of the contents pass through a
series of processes / steps that ensure a high quality of results.
Synset
It contains all information needed for the analysis of texts. It is the knowledge of the language in use, including proper
names, which can be expanded and customized in order to be aligned with particular ontologies.
Disambiguator
It distinguishes the different meanings of all the elements of a text by identifying the context in which they are located.
Categorization
It recognizes the categories covered in the text, assigning any idea to the correct element of the taxonomy.
Sentiment
It attributes a score to all occurrences identified in the text evaluation.
7
Customer Sentiment Monitor Process
Using Defined Drivers, a classification Thesaurus, domains and the
appropriate scheduling, the system indexes and collects all results
within a relational database for later use and analysis of findings.
Crawling
8
Analysis
Storage
Structured Data Results
Through the use of semantic
technologies, all the user
generated contents are
analyzed, understood and
evaluated for all defined
Drivers.
Each occurrence is
automatically given a vote.
Data are made available for
intelligence activities,
dashboard and reporting in
order to support the business
strategies.
9
Reporting
Through predefined templates or creation of personal representations,
the collected and evaluated data are aggregated in order to produce
the proper reports.
10
Example of Customer Sentiment Use
Product Lifecycle
IDEAS
R&D
IMPROVEMENTS
SATISFACTION
RISK & CRISIS
MANAGEMENT
Technical
Services
Consumer
Services
Legal &
Regulatory
CUSTOMER
INSIGHT
Sales &
Marketing
Customer Sentiment
Analysis
Monitor
UNSTRUCTURED
INFORMATION
11
SEMANTIC
ANALYSIS
STRUCTURED
DATA
Thanks
www.reply.eu