10:30-11:15am

 Track 1: SOCIAL MEDIA & BIG DATA
Moderator: Bryan Bell, Expert System
Topic: Voice of the Customer

Next Gen Voice of the Customer Using Text Analytics

The Voice of the Customer industry borrowed survey technology from the market research world because it was the only way to listen to the voice of the customer. Text analytics has changed all that, and just in time. The combined forces of consumer review sites, social media, and survey fatigue are ganging up to put consumers in the position of dictating the dialog between themselves and companies. Voice of the Customer programs can no longer simply rely on the venerable survey. They need to listen to the customer’s voice in their own words and wherever the customer shares.

Speaker: Kurt Williams, CTO, Mindshare Technologies

[ Top of this page ] [ Agenda overview ]

 10:30-11:15am

 Track 2: ENTERPRISE APPLICATIONS
Moderator: Meta Brown, LinguaSys
Topic: Knowledge Management

Enterprise Knowledge Management Via Text Search Analytics

This session will explore how Transomatic Corporation applied process to capture and manage content within the enterprise intranets, end user work stations and hard copy documents to provide pervasive text search and analytics. We’ll also discuss how a major labor organization manages 40 years of accumulated knowledge.

Speaker: Cameron Zadeh, Executive Director, Transomatic Corporation

[ Top of this page ] [ Agenda overview ]

 11:20-12:05pm

 Track 1: SOCIAL MEDIA & BIG DATA
Moderator: Bryan Bell, Expert System
Topic: Voice of the Customer

Content Intelligence – From What to Why Via Who, When and Where

We’ll discuss content Intelligence, how it adds value to Big Data, the Social Enterprise, Risk & Compliance, Collaboration and Distribution (basically any process where information in unstructured content needs to be brought into play).

This presentation includes use of a wide range of case studies and experience from real customer implementations. It features lessons learned, do’s and dont’s and shares proven return on investment evidence from various departmental and enterprise projects that Smart Logic customers have given permission to share.

Speaker: Jeremy Bentley, CEO, Smart Logic

[ Top of this page ] [ Agenda overview ]

 11:20-12:05pm

 Track 2: ENTERPRISE APPLICATIONS
Moderator: Meta Brown, LinguaSys
Topic: Knowledge Management

Industrial Knowledge Organization: Capturing and Leveraging Knowledge from Text

In the engineering services industry, field engineers routinely inspect, maintain and repair industrial machinery. The engineer’s observations and diagnosis are often captured in unstructured form in the maintenance logs, amounting to several hundreds of gigabytes. Technologies for mining these logs yield useful applications. For instance, a “tech assistant” application based on mining these logs assists field engineers in solving outstanding issues with the machinery, increasing workforce productivity. Eventually one can envisage “self-healing” systems wherein applications will automatically identify solutions to common recurring problems and apply the solutions automatically or with limited manual intervention.

Speaker: Steve Gustafson, Manager, Knowledge Discovery Lab, GE Global Research

[ Top of this page ] [ Agenda overview ]

 1:00-1:45pm

 Track 1: SOCIAL MEDIA & BIG DATA
Moderator: Bryan Bell, Expert System
Topic: Text Analytics & Politics

Social Networks Users Behaviors Analysis at the Revolution

Recently, Tunisia knew a pacific revolution, which had an impact on many Arabic countries. Tunisians demonstrated how to build a serene revolution without traditional ingredients such as icons and ideologies, and showed how to start a real democracy. This serene revolution was a result of the rising of a relatively small, homogenous, and educated population aiming for democracy and for a better life. In this work, we are interested in the analysis of the impact of this revolution on the behavior of Tunisian people. For this purpose, we perform two main tasks: information collection, and post texts analysis and mining.

Speaker: Jalel Akaichi , Chair, Master’s of Science in Business Intelligence Program, University of Tunis

[ Top of this page ] [ Agenda overview ]

 1:00-1:45pm

 Track 2: ENTERPRISE APPLICATIONS
Moderator: Meta Brown, LinguaSys
Topic: Search

Is Big Data Giving Us The Answer?

While full-text search worked with large data collections for some time, recent activity summarized as “Big Data” might represent a paradigm shift. While amassed data can yield many answers, a new question arises: do we need big data to query Big Data?

Speaker: Martin Baumgartel, Site Search Manager, Walt Disney

[ Top of this page ] [ Agenda overview ]

 1:50-2:35pm

 Track 1: SOCIAL MEDIA & BIG DATA
Moderator: Bryan Bell, Expert System
Topic: Text Analytics & Politics

Semantic Analysis of the Political Discourse in the Presidential and Congressional Campaigns of 2012

The federal election campaign of 2012 is generally thought to be negative and to present challenges as we try to construct a new political reality for the future. The research reported semantically analyzes the political discourse of the 2012 Presidential and Congressional campaigns, focusing on several prominent candidates and political commentators. Semantic analysis integrates well established factors from the psychiatric and psychological literature, including anxiety, fear, hostility, aggressiveness, depression, happiness, vision, and contentedness. The results are compared to subjective human analyses of the campaign. Results highlight the value of objective analytical methods and validated professional models.

Speaker: Denise Bedford, Goodyear Professor of Knowledge Management, College of Communication and Information, Kent State University

[ Top of this page ] [ Agenda overview ]

 1:50-2:35pm

 Track 2: ENTERPRISE APPLICATIONS
Moderator: Meta Brown, LinguaSys
Topic: Search

Beehive: Semantic Search Optimization at Autodesk

Project Beehive’s objective is to deliver highly relevant content about how to use our products directly within the software, at point of need. As a part of this effort, the project initially has focused on enabling and optimizing search across numerous content silos leveraging ontologies to improve relevancy and to aggregate related content. The project has been scaled to accommodate all customer facing content on Autodesk’s web sites, processing tens of millions of queries on a yearly basis. This presentation will discuss the business case for the project, technical infrastructure, the text analytics and ontologies.

Speakers:
Tom Williams, Director – Experience & Engineering, Autodesk Knowledge Network, Autodesk
Shamik Bandopadhyay, Senior Search/Community Web Engineer, Autodesk
Michael Kilgore, Principal, InfoClear
Sunny Yoon, Taxonomist, Autodesk

[ Top of this page ] [ Agenda overview ]

 3:00-3:45pm

 Track 1: SOCIAL MEDIA & BIG DATA
Moderator: Bryan Bell, Expert System
Topic: Advanced Text Analytics Techniques

How to Evaluate Text Analytics Software

Session description coming soon!

Speaker: Tom Reamy, Conference Chair, Text Analytics World

[ Top of this page ] [ Agenda overview ]

 3:00-3:45pm

 Track 2: ENTERPRISE APPLICATIONS
Moderator: Meta Brown, LinguaSys
Topic: New Applications

TBD

Session description is coming soon!

Speaker: Leslie Owens, Principal Analyst & Research Director, Forrester Research, Inc.

[ Top of this page ] [ Agenda overview ]

 3:50-4:35pm

 Track 1: SOCIAL MEDIA & BIG DATA
Moderator: Bryan Bell, Expert System
Topic: Advanced Text Analytics Techniques

Sentiment Trend Intelligence: Getting Ready for the Future

Trend emergence analytics remains a frontier practice area for social marketing and sentiment analysis. At the same time, a solid understanding of impactful conversation can make the difference between preparedness and disaster for the enterprise. In this session, we will discuss the issues, technologies, and best practices for discovering new themes, providing a robust set of granular analytics around their social features, and delivering these patterns to a decision framework for social media.

Speaker: Hadley Reynolds, Managing Director, NextEra Research

[ Top of this page ] [ Agenda overview ]

 3:50-4:35pm

 Track 2: ENTERPRISE APPLICATIONS
Moderator: Meta Brown, LinguaSys
Topic: New Applications

E-Discovery Through Text Mining: Fraud Detection

Large fraud investigations cases present require analyzing huge amounts of textual data with a very vaguely defined objective. Investigators need to process a haystack, but they are not even sure what they need to be looking for: this might be a needle, but can be some other small thing as well. This case study outlines key tasks and challenges a government agency had to address when looking for potential signatures of fraud in a large collection of subpoenaed corporate emails. We will outline a sequence of steps that lead to focusing attention of investigators on several thousand most relevant messages.

Speaker: Sergei Ananyan, CEO, Megaputer Intelligence

[ Top of this page ] [ Agenda overview ]

 Track 1: SOCIAL MEDIA / BENEFITS OF TEXT ANALYTICS
Moderator: Andrew Fast, Elder Research, Inc.
Topic: Text Analytics Applications

Picture Perfect: Leveraging Text Analytics to Suggest Stock Art for Web & Print Content

While designing Web or print publications, there is a typical requirement to place stock art at regular intervals to keep up reader engagement and break the monotony of text. Selecting relevant stock images is time-consuming and often involves collaboration between multiple individuals. There are copyright considerations as well: you cannot simply select any image that you come across. The requirement to maintain attribution for the images in use poses a challenge as well.

This presentation examines the use of natural language processing—particularly named-entity recognition (NER) techniques—in suggesting context-aware, free-to-use, and well-attributed stock images for Web or print content.

Speaker: Diana Joseph, Learning Strategist, Adobe Systems

[ Top of this page ] [ Agenda overview ]

 10:40-11:25am

 Track 2: KNOWLEDGE ORGANIZATION
Moderator: Tom Reamy, Text Analytics World
Topic: Ontology-Based Applications

AllegroText: Text Mining with an RDF Graph Database

AllegroText is a framework that includes a web and news crawler, an automatic link to various entity extractors, an enrichment module that uses the Linked Open Data cloud and a scalable graph database. With this framework, one can track entities like people, products or organizations and provide detailed semantic and statistical analytics. We’ll discuss a practical application that detects risk in the supply chain of car manufacturers by finding how natural disasters or political unrest might influence your supply chain.

Speaker: Jans Aasman, President and CEO, Franz Inc.

[ Top of this page ] [ Agenda overview ]

 11:30-12:15pm

 Track 1: SOCIAL MEDIA / BENEFITS OF TEXT ANALYTICS
Moderator: Andrew Fast, Elder Research, Inc.
Topic: Conversations

Supporting Two-Way Customer Interaction with Synergistic Text Analytics and Taxonomy and Metadata Strategies

The ability to engage customers in a 2-way conversation has transformed many business environments. Successfully navigating and utilizing these conversations requires that organizations understand what customers are saying individually as well as in the aggregate. Text Analytics is critical to attaining this understanding, which must be rapidly leveraged to enable business response. A technical content strategy involving metadata and taxonomies supports this requirement by enabling content to be quickly deployed or reorganized to drive the conversation. In this talk, we expand on the benefits of utilizing text analytics in conjunction with taxonomies and metadata to directly support your customer interaction.

Speaker: Carol Hert, Senior Consultant, Factor

[ Top of this page ] [ Agenda overview ]

 11:30-12:15pm

 Track 2: KNOWLEDGE ORGANIZATION
Moderator: Tom Reamy, Text Analytics World
Topic: Auto-Categorization

From Print-First to Digital-Centric B2B Media with Automatic Categorization

Hanley Wood has overhauled and streamlined their workflow from a print-first to digital-centric model using dynamic, metadata-driven websites integrated with data products, all with an increasingly lean staff. Automatic categorization, standardized controlled vocabularies, and categorization-driven aggregation have enabled organizational innovation at this a business-to-business media company that publishes more than 30 brands.

Speaker: Monica Fulvio , Senior Librarian, Content Architecture, Hanley Wood

[ Top of this page ] [ Agenda overview ]

 1:15-2:00pm

 Track 1: SOCIAL MEDIA / BENEFITS OF TEXT ANALYTICS
Moderator: Andrew Fast, Elder Research, Inc.
Topic: Conversations

Using Context for Deep Text Analytics

When people communicate online, they use their knowledge of the world to talk creatively about everything from products to news. Gaining meaning from this text means going beyond set keywords and ontologies to a deep understanding of most of the explicit and implicit ways of communicating. We explain how concept-based text understanding solves this problem by using models of how people think about the world to reason about language without the use of rules, lexicons, training, or ontologies. We discuss how common sense techniques which model how people think of the world can add context to text understanding.

Speaker: Catherine Havasi, CEO, Luminoso

[ Top of this page ] [ Agenda overview ]

 1:15-2:00pm

 Track 2: KNOWLEDGE ORGANIZATION
Moderator: Tom Reamy, Text Analytics World
Topic: Auto-Categorization

Transparent Rule-Based Systems for Technology Assisted Review

Electronic Discovery has seen an explosion of interest in using Text Analytics technologies for Technology Assisted Review (TAR) with substantial savings in legal expenditures for enterprises. While much attention has been focused on machine learning based approaches to TAR, a number of other TAR methodologies show promise. Rules-based approaches, in particular, offer an alternative that is highly accurate while remaining simple, transparent, and defensible. A panel of experts will explain how rules-based technology assisted review differs from the usual paradigm and how it is used to support highly accurate, defensible document classification.

Speakers:
Bruce Hedin, Principal Scientist, H5
Chris Hogan, Head of Advanced Research and Development, H5

[ Top of this page ] [ Agenda overview ]

 2:05-2:50pm

 Track 1: SOCIAL MEDIA / BENEFITS OF TEXT ANALYTICS
Moderator: Andrew Fast, Elder Research, Inc.
Topic: Benefits of Text Analytics

Making Text Analytics Pay

You can deliver accurate analysis of customer sentiment. So what?
Your word cloud reflects brand image among twitter users. So what?
Your interactive graphic dashboard enables exploration of concepts. So what?

Executives don’t want to follow the dots between your technology and their bank accounts. They want clear connections between what you can do and cold, hard cash.

In this session, you will learn:

• Why most current text analytics applications don’t mean squat to executives.
• What buttons to push when pitching text analytics to executives.
• How to identify text analytics applications that are good bets for strong ROI.

Speaker: Meta Brown, Independent Consultant

[ Top of this page ] [ Agenda overview ]

 2:05-2:50pm

 Track 2: KNOWLEDGE ORGANIZATION
Moderator: Tom Reamy, Text Analytics World
Topic: Taxonomy and Text Analytics

Automatic Taxonomy Generation for a News Group

Creating and maintaining taxonomies for the news domain is a tedious task. Human experts spend many hours constructing taxonomies manually and generating rules for assigning taxonomy concepts to newly published articles.

What if a taxonomy was generated automatically straight from the news articles? One wouldn’t have to worry about new terms and concepts appearing in the news, or categorizing the articles.

In this talk, Anna Divoli will present a case study on generating a taxonomy from articles published by the FairFax Media NZ. Pingar utilized text analytics techniques and Linked Data sources to create a taxonomy focused on the current news.

Speaker: Anna Divoli, Senior Software Researcher, Pingar

[ Top of this page ] [ Agenda overview ]

 2:05-2:50pm

 Track 2: KNOWLEDGE ORGANIZATION
Moderator: Tom Reamy, Text Analytics World
Topic: Taxonomy and Text Analytics

Capturing Acronyms for a Medical Taxonomy

Acronyms occupy a special place in medical literature, and any medical taxonomy must include them. However, there are literally tens of thousands of acronyms, many with dozens of meanings. How do we find acronyms, how do we decide which ones we want, how do we keep them straight once we have them, and how do we make them available to applications? Find out how acronyms are handled for EMMeT, Elsevier’s Merged Medical Taxonomy, to make our applications more intuitive to users

Speaker: Douglas Anderson, Knowledge Classification Specialist, Elsevier

[ Top of this page ] [ Agenda overview ]

 3:10-3:55pm

 Track 1: SOCIAL MEDIA / BENEFITS OF TEXT ANALYTICS
Moderator: Andrew Fast, Elder Research, Inc.
Topic: Taxonomy and Text Analytics

Getting Started With Business Taxonomy – Unlocking the Keys to Successful Information Management

Business users are more active than ever in the publication and consumption of information. From big data, to social computing, to content and document management, more information is being generated and thus the need to effectively manage and retrieve that information is all the more critical. Business taxonomies present an integral component of this information management solution. This presentation will provide best practices and a proven approach for the design and development of intuitive and sustainable taxonomies that will serve today’s information management efforts and will offer a series of real world examples from the public and private sectors.

Speaker: Zach Wahl, Vice President, Project Performance Corporation

[ Top of this page ] [ Agenda overview ]

 3:10-3:55pm

 Track 2: KNOWLEDGE ORGANIZATION
Moderator: Tom Reamy, Text Analytics World
Topic: Taxonomy and Text Analytics

Semantic Metadata’s Two Types of Vocabularies

Semantic metadata is metadata that is expressed using a standard syntax that can be commonly processed by applications and tools. This talk will focus on: the two types of vocabularies involved with semantic metadata, the class-attribute vocabulary, and the value vocabulary. We will explain why the most common vocabularies in the linked data cloud are Dublin Core, FOAF and SKOS; and how canonical lists of named entities (people, organizations, places, events and things) are being used to power new semantic services on web search engines, news websites and online shopping.

Speaker: Joseph Busch, Founder, Taxonomy Strategies

[ Top of this page ] [ Agenda overview ]