=Paper=
{{Paper
|id=None
|storemode=property
|title=T-OMT: A Novel Opinion Mining Tool for Improving Global Customer Relationship Management
|pdfUrl=https://ceur-ws.org/Vol-1013/casfe2013_submission_5.pdf
|volume=Vol-1013
|dblpUrl=https://dblp.org/rec/conf/casfe/Rodriguez-GancedoCRP13
}}
==T-OMT: A Novel Opinion Mining Tool for Improving Global Customer Relationship Management==
https://ceur-ws.org/Vol-1013/casfe2013_submission_5.pdf
T-OMT: A Novel Opinion Mining Tool for Improving Global
Customer Relationship Management
Mari Carmen Rodríguez-Gancedo1, Javier Caminero1, José Relaño1, Carlos Picazo2
1
Telefónica R&D, Madrid, Spain; 2Polytechnic University of Madrid, Spain
Email: {mcrg, fjcg}@tid.es
ABSTRACT different sources to express opinion must be taken into
For IT companies, a quick reaction when a new product is account for having an unbiased view of any situation.
being deployed into the market is crucial, so a very Telefonica count with a corporate customer portal, where
dynamic response is required, mainly analyzing the new users are able to access all kind of information related with
channels used by their customers to express their company products and services and also to make
satisfaction or concerns. This quick response can only be comments, suggestions or complaints about them, by means
achieved if companies count with the support of powerful of emails that will be treated by very skilled operators. For
Opinion Mining Tools able to analyze the sentiment information exchange about company’s products and
evolution over the new channels like social networks or services, the customer portal also has independent blog and
corporate blogs and forums, then these have become the forums.
most common choice for obtaining a valuable feedback and
For any global company, it is important to address and
traditional channels like filling out forms or call centers
anticipate user’s needs, and that can only be achieved by
have been relegated to a secondary position. In the present
analyzing the different interactions from the diverse
paper, we present a novel and powerful tool for opinion
channels where users express their concerns or suggestions
mining developed by Telefonica R&D in the framework of
about new products and services. Given the global scope of
the Render project, able to produce quick reports with user-
the company, a detailed analysis of the different markets
friendly charts to analyze information coming from
where the company is present is a must. Specifically,
different sources and providing the company with a
Telefonica is present in 24 countries and sometimes
powerful weapon to be able to react to negative sentiment
product cannot be directly launched into the different
when releasing new products or services, and adapting
countries and some adaptations are needed taken into
them to the needs of the customers, that will actually be the
account the idiosyncrasy of people.
final beneficiaries, then the products and services will be
much more adapted to their currents needs. Analyzing the available information can be quite
challenging, then in some cases is somewhat structured,
INTRODUCTION & MOTIVATION
and in many others free text is usually the norm in the
Nowadays, companies involved in a global and dynamic
contacts found in open forums or social media.
market should be very careful with the needs of their actual
customers to keep their loyalty. Therefore, new products to RENDER PROJECT FRAMEWORK
be launched must always take into account the current The main motivation for Render project 1 starts from the
needs of the customers and the community related with unprecedented success of the Web for facilitating the
them in order to be able to attract new customers with publication, use and exchange of information on a global
similar needs. scale, on virtually every topic, and representing an amazing
diversity of opinions, viewpoints, mindsets and
Nowadays, information changes very rapidly and social
backgrounds. Its design principles and core technological
networks have become customer’s preferred channel to
components have led to an unprecedented growth and mass
share opinions, knowledge, advice and dissatisfaction,
collaboration. This trend is also finding increasing adoption
although for global companies like Telefonica with
in business environments. Nevertheless, the Web is also
heterogeneous markets and customers, some customers still
confronted with fundamental challenges with respect to the
continue using the traditional channels like call centers, or
purposeful access, processing and management of these
personally visiting company stores, so an integration of the
sheer amounts of information, whilst remaining true to its
principles, and leveraging the diversity inherently unfolding
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
through world wide scale collaboration.
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. To copy otherwise,
or republish, to post on servers or to redistribute to lists, requires prior 1
http://www.render-project.eu/
specific permission and/or a fee.
Render engages with these challenges by developing comments, discussion forums and customer feedback
methods, techniques, software and data sets that leverages databases from Telefonica, all together adding up to
diversity as a crucial source of innovation and creativity, hundreds of millions of items, some even on a daily
whilst providing enhanced support for feasibly managing basis. This data can be managed by a highly scalable
data at very large scale, and for designing novel algorithms data management infrastructure, and enriched with
that reflect diversity in the ways information is selected, machine-understandable descriptions and links
ranked, aggregated, presented and used. Render’s referring to the Linked Open Data Cloud.
information management solution scales not only to large • Identify and extract the diversity embodied within the
amounts of data and users, but also to a plurality of points various information sources collected, and make the
of views and opinions. This has been demonstrated through connections and references between different items and
the usage of realistic data sources, including news streams sources explicit.
covering over 5,000 sources worldwide with 100,000 items • Represent and process diversely expressed information
per day, (micro) blog streams adding up to more than a so as to explicate and conceptualize the results of the
million posts per day, a full data stream from Wikipedia, mining task, to enable the development of diversified
and the Linked Open Data Cloud; through open source information management algorithms and services.
extensions to popular collaboration and communication Render is developing novel, scalable techniques to
platforms such as MediaWiki and Drupal; and through reason upon opinions and viewpoints, and for
three high-profile case studies. diversity-aware information selection and ranking.
Render is helping to realize a world where information is Render also looks into proper means to make diversity
acquired and shared in a fundamentally different manner information accessible to the end-user by providing
than the consensual approach promoted by movements such sophisticated metaphors, interfaces and software tools
as Web 2.0, and where communication and collaboration to organize, display and visualize it. Raking algorithms
across the borders of social, cultural or professional take into account the viewpoints underlying different
communities are truly enabled via advanced Web information items.
technology, supporting one of the credos of European • Use diversity as integral concept of popular
society: “United in diversity”. communication, collaboration and information sharing
platforms, in the form of extensions to MediaWiki,
Drupal, or Twitter. Render technology allows to
The Approach explicitly link to items with a dissenting view, and thus
Render provides a comprehensive conceptual framework to increase the diversity exposure of the wider Web
and technological infrastructure for enabling, supporting, audience.
managing and exploiting information diversity in Web-
based environments.
T-OMT, TELEFONICA OPINION MINING TOOL
Diversity is a crucial source of innovation and adaptability. Framework
It ensures the availability of alternative approaches towards
Given the already mentioned difficulties and specific goals
solving hard problems, and provides new perspectives and
of IT companies for dealing with user’s opinion and the
insights on known situations.
mandatory quick responses times required, a new tool has
Equally important, embracing diversity in information been designed and implemented based on the Render
management is essential for enhancing state-of-the-art project approach. This tool called T-OMT, which stands for
technology in this field with novel paradigms, models, and Telefonica Opinion Mining Tool, constitutes one of the
methods and techniques for searching, selecting, ranking, main use cases and outcomes of the project.
aggregating, clustering and presenting information
Large companies that provide several products and services
purposefully to users, thus alleviating critical aspects of
need customer’s feedback about their products and its
information overload.
impact on a global market with hundreds of millions of
Render is developing concepts, methods, techniques and users. The customers publish their opinions about these
technology to: products in different media (Twitter, email, surveys…)
• Collect and manage information sources rich in every day. The goal of T-OMT has been to collect massive
diversity so that this information is available in an datasets from these sources and use Render processes,
effective form and can be processed efficiently in giving the user analytic and useful information about the
further steps. It is done by crawling, gathering, current opinion of the product. This tool uses Twitter as a
structuring and enriching various information sources primary opinion source and generates specific reports
with a great diversity basis, including sources relevant filtering the data by date, topics and language. These
for the Render case studies. Render leverages very reports are showed also graphically, detailing the mentions
large amounts of content and metadata: news, blog and of the topics, global sentiment, opinion geolocation, and
microblog streams, content and logs from Wikipedia, details about the related topics.
news archives, multimedia content and reader
T-OMT has been designed to satisfy the detected needs of After discarding some other possibilities and having into
the final users, allowing the analysis of several data sources account that the front-end of the application will have a
provided by communications channels with customers and web interface and it will require some data processing,
also potential customers. HTML5 is the chosen technology to build the interfaces
T-OMT’s stakeholders are the final users of Telefonica’s and for the definition of the graphical aspects, CSS3 is
corporate portal where the Render opinion mining used, since it is considered a standard for the design of web
capabilities are being deployed. The final users are mainly interfaces and finally, for the data processing of the front-
enterprises or departments both inside Telefonica group and end, JavaScript, adding a jQuery library has been the
also in another third party service provided designing choice, for its simplicity, robustness and multidevice
applications for final customers focused mainly on these compatibility and able to deal with JSON objects, that are
groups: the outputs provided by the back-end. Additional software
components and plugins are used, mainly for graphical
• Social media marketing, providing mainly social media widgets. GeoMap and line/pie charts are powered by
profiles of the company. The main functions are: Google charts libraries and their online technologies.
o Community Management.
o Evaluation of the impact of new products/services To be consistent with the main libraries and functionalities
or advertisement campaigns. jQuery-ui is the library included for graphical web building.
o Influence level of different online media. Its web browser compatibility, cross-device design and
o Competition products analysis. robustness make this plugin highly properly for this tool.
• Corporate Reputation, checking between several The server side is developed in Java6 and deployed in an
issues, the online reputation of the company. The Apache-Tomcat server. A ReSTful API implementing the
online reputation is centered in: real-time required functionality over AJAX methods is used
o Detection of negative opinions to allow an early as interface between the front-end and the server.
correction. The application databases use MongoDB technology. It
o Detection of brand perception and knowledge stores user information and processed reports saved by the
level in Internet. user to be loaded without reprocessing the data in order to
o Analysis of attributes associated to the brand. reduce response times and enhancing backend’s scalability.
• Business Intelligence, managing the extraction of
knowledge through analysis of existing data in a
company. Main Toolkit Features
In the first place, a panel to filter the information to be
treated is required. This panel will allow the application to
Architecture and Implementation Details generate reports according to the restrictions inserted by the
The Opinion Mining Tool consists of two main user and also recovering previously saved reports, avoiding
components, a backend for the data statistical analysis and thus, the creation of repeated reports from the system.
a frontend as user interface. The different filters make it possible to make a report
The architecture of components is displayed in Figure 1, generation according to the following criteria:
showing the different components for performing queries
• User Group: Online / Offline
and displaying the right graphical charts as a result of the
analysis. • Source Type: Twitter, Call-Center, Survey …
• Country: Spain, Germany, UK ...
• Topic: individual or multiple topics can be added
• Time Period: a timeframe can be selected
Generated Reports Details
The generated report is shown through two main panels
where all the information is shown (see Figures 2 and 3):
The first panel (Figure 2) is the “GeoMap” panel, showing
a world map detailing the number of topic found in the
different countries. On the top area of the panel three
different charts are also displayed and that can provide with
additional information by clicking on them:
• A ‘mention chart’, showing the number of mentions of
Figure 1: T-OMT Architecture diagram a topic over the time (different topics can be displayed
simultaneously). In a new product launch, for example, • A ‘related-topic word cloud’, showing the most
this chart can help quickly visualize the evolution of frequent appearances in a bigger size and the less
the mentions of the product over the time. In this chart, frequent in a smaller one.
the sentiment is not taken into account.
• An ‘overall sentiment’ chart, displaying graphically the
positive and negative sentiment. CONCLUSIONS
In the framework of Customer Relationship Management, it
• A ‘geo-distribution’ chart, showing the geographical
is nowadays mandatory to have proper tools to be able to
proportion of appearances of the selected topics
analyze and give proper quick responses to the needs of
customers in a global and competitive world, where
opinions and sentiment evolve very fast and new
communication channels like the social networks are
quickly replacing traditional channels like call centers.
In response to these needs, Render project has successfully
provide a novel approach, and focused its efforts mainly in
the new interaction channels, but without forgetting the
traditional ones, then for global companies, in certain
countries, traditional channels may still play an important
role.
A new Opinion Mining Tool has been presented by
Telefonica as a project use case and one of its major
outcomes. This tool is able to generate reports analyzing
massive information coming from different channels,
Figure 2: ‘GeoMap’ Graphical Output Panel
including social networks and it shows valuable charts as a
result of a deep analysis of the data and providing details as
The second panel (Figure 3) shows details of the most the evolution of the sentiment over the time and its
frequent topics, and at the top the same three charts geographical distribution, a useful and powerful tool
displayed on the previous panel are also shown. particularly useful for global companies like Telefonica.
Thanks to the results provided by the tool, return of
investment can be enhanced, missed business opportunities
will be avoided and opinions, viewpoints, and customers
ideas can be turned into a competitive advantage.
ACKNOWLEDGEMENTS
This work received funding from the European
Commission’s Seventh Framework Program under grant
agreement number 257790 (FP7-ICT-2009-5).
REFERENCES
1. T. O’Reilly. “What is Web 2.0?” Design Patterns and
Figure 3: Details Output Panel Business Models for the Next Generation of Software”, Sep.
2005. http://oreilly.com/web2/archive/what-is-web-20.html
2. Ha He, Hugo Haas and David Orchard. “Web Services
Additionally to the showed charts, some more could be Architecture Usage Scenarios”. Technical report - World
interesting to be added: Wide Web Consortium (W3C), 2004.
• A ‘sentiment evolution chart’, showing the evolution 3. M. Damova, K. Simov, Z. Tashev, and A. Kiryakov,
of the sentiment of a topic over the time (different "FactForge: Data Service or Diversity through Inferred
topics can be displayed simultaneously). Knowledge over LOD," in Proceedings of AIMSA’2012,
• A ‘detailed geo-distribution chart’, analyzing how the Bulgaria, 2012.
appearances grow up or decrease in a certain 4. A. Thalhammer, I. Toma, R. Hasan, E. Simperl, and D.
Vrandecic, “How to Represent Knowledge Diversity”, 10th
geographical area.
International Semantic Web Conference ISWC’11, Germany,
2011.