Track 1: SOCIAL MEDIA / SENTIMENT

Track 2: ENTERPRISE TEXT ANALYSIS

Monday, March 17, 2014

8:00-9:00am
Room: Grand Assembly

Registration & Networking Breakfast


9:00-9:10am
Room: Golden Gate C2

Conference Chair Welcome Remarks

Speaker: Tom Reamy, Chief Knowledge Architect, KAPS Group

[ Top of this page ] [ Agenda overview ]


9:10-10:00am
Room: Golden Gate C2

Keynote
Enhancing Search with Predictive Analytics

Search is the cornerstone of enterprise information management. As the amount of searchable information grows, however, keyword relevance becomes increasingly inadequate for finding valuable information. In this talk, we highlight successes we’ve had incorporating results from predictive analytics models into search, including document similarity, predictive facets and enhanced relevance functions.

Speaker: Andrew Fast, Chief Scientist, Elder Research, Inc.

[ Top of this page ] [ Agenda overview ]


10:00-10:30am
Room: Salon 8 & 9

Exhibits & Morning Coffee Break

[ Top of this page ] [ Agenda overview ]


10:30-11:15am
Room: Golden Gate C1

Track 1: SOCIAL MEDIA / SENTIMENT
Topic: New Techniques & Approaches

Social Media Alerts Using Semi-Supervised Learning

Most social media monitoring systems today utilize explicit keywords in Boolean expressions to create alerts – a manually intensive and inherently fragile process. In this session, we describe an alternative approach for creating relevant alerts, utilizing high-level topic-models constructed through semi-supervised learning methods. The input to the system is simply a set of documents pertaining to the business, which the system analyses with input from the user to create “soft” filter criteria for relevance estimation. This approach, coupled with supervised learning for filter refinement, has been implemented and proven to provide a more relevant filtered-alert stream with minimal maintenance.

Speakers:
Pradeep Sinha, CEO / CTO, Contata Solutions

Tony Schulte, Algorithms Researcher, Contata Solutions

[ Top of this page ] [ Agenda overview ]


10:30-11:15am
Room: Golden Gate C2

Track 2: ENTERPRISE TEXT ANALYSIS
Topic: New Applications

Lessons Learned from the Analysis of K to 12 Disciplinary Records to Minimize Expulsion Rates

Today’s K to 12 public educational system is seeing a growth in discipline problems among children of all ages. The growth of these problems diminishes resources and time on tasks for learning. By looking at a collection of records with a narrative explaining the event in detail, it was possible to see trends and patterns. The unstructured data describing the infraction yielded the true source of the problem. This session focuses on what was done — and what should have been done — to produce better outcomes.

Speaker: Chris Bowman, President, Educational Analytics and Logistics

[ Top of this page ] [ Agenda overview ]


11:20-12:05pm
Room: Golden Gate C1

Track 1: SOCIAL MEDIA / SENTIMENT
Topic: New Techniques & Approaches

Normalizing Sentiment in Multilingual Contexts

Sentiment Analysis, or Opinion Mining, extracts insights from an unstructured data source and demonstrates the computational treatment of opinion, sentiment and subjectivity. Many of the resources developed for this type of information extraction are in English and designed for this language. Some would agree that the accuracy of a popular solution, machine translation, is questionable. This presentation from explores various methodologies for normalizing sentiment analysis with respect to multilingual resources and across English varieties.

Speakers:
Durjoy Patranabish, Senior Vice President, blueocean market intelligence
Payel Dutta Chowdhury, Data Scientist, blueocean market intelligence

[ Top of this page ] [ Agenda overview ]


11:20-12:05pm
Room: Golden Gate C2

Track 2: ENTERPRISE TEXT ANALYSIS
Topic: New Applications

Real-Time Monitoring for Corporate Risk Mitigation

With today’s deluge of email, text, and documents, Corporate Management has nearly zero visibility and control of contractual risk. Today’s current technologies essentially work “after the fact” to contain financial damage once an issue exists. In this presentation, we discuss graph-based technology solutions that facilitate the linking of relevant content (such as email) to important contract terms in order to anticipate and quantify potential legal risks.

Decomposing complex relationships, via graph technologies, into discrete components and linking these components to relevant content — inside and outside of a corporation — promises to facilitate real-time calculation of risk.

Speaker: Jans Aasman, CEO, Franz Inc.

[ Top of this page ] [ Agenda overview ]


12:05-1:30pm
Room: Salon 8 & 9

Lunch

[ Top of this page ] [ Agenda overview ]


1:30-2:15pm
Room: Golden Gate C2

Expert Panel

Moderator: Tom Reamy, Chief Knowledge Architect, KAPS Group, LLC

Speakers:
Bryan Bell, Vice President of Global Marketing, Expert System
Christine Connors, Founder & Information Strategist, TriviumRLG
Antonio Valderrabanos, CEO, Bitext – The Bits and Text Company

[ Top of this page ] [ Agenda overview ]


2:15-2:30pm
Room: Golden Gate C2

Sponsor Presentation:
Using Social Data to Forecast Key Performance Indicators

Gnip

Pivotal built a distributed topic & sentiment analysis pipeline using our Greenplum Database, open source tools like MADlib and PL/Python and Twitter feeds from Gnip. We’ll discuss how Pivotal used this pipeline to build regression models to predict commodity futures from tweets for one major agri-business client and to enhance churn models for a telecom client. With the ease of data access through Gnip and the power of the Pivotal platform, we were able to build this pipeline and deliver results to our customer in less than 2 weeks.

Speaker: Srivatsan Ramanujam, Senior Data Scientist, Pivotal

[ Top of this page ] [ Agenda overview ]


2:35-3:15pm
Room: Golden Gate C1

Track 1: SOCIAL MEDIA / SENTIMENT
Topic: New Techniques & Approaches

Case Study: Canadian Imperial Bank of Commerce
Discovering Patterns in Client Feedback

Most organizations have taken initiative to derive sentiment and categories from their client feedback. These efforts summarize conversations into meaningful groups that expose insight. As feedback grows and time passes, this level of summarization becomes insufficient. Among thousands of categories, patterns of co-occurrence are bound to occur that will be missed by the human analyst. CIBC has developed a customized process of Sequence Discovery, Decision Trees and Visualization techniques to build a Voice-of-Client analysis strategy that goes beyond deriving sentiment and categories to discover patterns. The approach is simple, yet sophisticated, and scalable to the enterprise.

Speaker: Masoud Charkhabi, Director, Advanced Analytics, Canadian Imperial Bank of Commerce

[ Top of this page ] [ Agenda overview ]


2:35-3:15pm
Room: Golden Gate C2

Track 2: ENTERPRISE TEXT ANALYSIS
Topic: New Applications

Case Study: A Fortune 500 Company
Using Text Analytics to Improve the User Experience

When reviewing analytics, it’s easy to see when people are coming, where they’re coming from, and when they’re leaving. But understanding why they leave remains a big challenge for marketers and analysts. This session examines that concern by analyzing online surveys as a way to gain a more in-depth understanding of content from users. Learn how a Fortune 500 company leveraged their quantitative data and deployed a survey to glean more insight on the qualitative data and determine where the faults existed. Explore how to get more meaningful data and how to provide a better overall user experience.

Speaker: Feras Alhlou, Principal Consultant/Co-Founder, E-Nor

[ Top of this page ] [ Agenda overview ]


3:15-3:50pm
Room: Salon 8 & 9

Exhibits & Afternoon Break

[ Top of this page ] [ Agenda overview ]


3:55-4:40pm
Room: Golden Gate C1

Track 1: SOCIAL MEDIA / SENTIMENT
Topic: New Techniques & Approaches

Case Study: Johnson Space Center
Usability/Sentiment for the Enterprise and ENTERPRISE

THE solution for indexing, classifying and transferring knowledge is difficult to identify with the dynamic nature of software components, metadata application, and interface options. While the perceived value of content is subjective to the perceptions of the end user, the complexity of the infrastructure and operation provided by IT can overshadow the need for seamless delivery. This presentation (and discussion!) focuses on a bi-layered measure of usability and sentiment analysis for a local and national audience of scientists, researchers and engineers.

Speaker: Sarah Ann Berndt, Taxonomist, Johnson Space Center
David Meza, Operational Excellence Program Manager, Johnson Space Center

[ Top of this page ] [ Agenda overview ]


4:45-5:30pm
Room: Golden Gate C1

Track 1: SOCIAL MEDIA / SENTIMENT
Topic: New Techniques & Approaches

Beyond Sentiment: Understanding Customers through Experience Modeling

When we reduce a tweet, a post, or any social message to a single positive/negative data point, we do violence to the complexity of our customers’ expressions. Using simple sentiment analysis to understand the customer experience is no better than using black-and-white photographs to understand color.

This session introduces methods for modeling the emotional and conceptual resonances in social media data by exploiting the expression-centric structure of corpora like the Experience Project. Through speaker-sensitive word association modeling, we can embrace the complexity of language without compromising the perspicuity of our results.

Speaker: Moritz Sudhof, Algorithmic Lead, Kanjoya

[ Top of this page ] [ Agenda overview ]


4:45-5:30pm
Room: Golden Gate C2

Track 2: ENTERPRISE TEXT ANALYSIS
Topic: New Applications

Case Study: Luminoso
Market Research: Leaving the Uncanny Valley

Your email inbox is full of customer complaints, product suggestions, and feedback. Your surveys are full of open-ended questions that surely contain good ideas and things that need work. But how do you find the time to read even a tenth of all that data?

Understanding human-directed language requires a deep understanding that goes beyond connections between words into the ever-evolving way that we view the world. This session examines a case study in market analytics, supply chain management, and product development to show how deeper text understanding can provide insights that are immediately useful!

Speaker: Catherine Havasi, CEO, Luminoso

[ Top of this page ] [ Agenda overview ]



5:30-7:30pm
Room: Salon 8 & 9

Networking Reception

[ Top of this page ] [ Agenda overview ]


Track 1: SOCIAL MEDIA / SENTIMENT

Track 2: ENTERPRISE TEXT ANALYSIS

Tuesday, March 18, 2014

8:00-8:45am
Room: Grand Assembly

Registration


8:00-8:45am
Room: Salon 8 & 9

Networking Breakfast


8:45-9:00am
Room: Golden Gate C2

Conference Chair Welcome Remarks

Speaker: Tom Reamy, Chief Knowledge Architect, KAPS Group


9:00-9:15am
Room: Golden Gate C2

Sponsor Presentation:
Mountains of Data: Intelligent Knowledge Management for the 21st Century

Kcomber

With the rapid increase in the amount of data organizations have to manage, technology becomes an essential tool in providing quality Business Intelligence. Aiming to take information management to a new level, Kcomber, a knowledge management company, is committed to providing Intelligent Knowledge Management by automating data collection, processing, analysis and modeling to assist our clients in gaining a competitive advantage in the world of “big data” management. In this session, the speaker will discuss problems and solutions in Big Data and Knowledge Management as well as descriptions of the techniques, methodologies and solutions used by Kcomber Inc. in some of its projects.

Speaker: Mr. Zhonghui Wu Founder and Managing Director, Kcomber Inc.

[ Top of this page ] [ Agenda overview ]


9:15-10:00am
Room: Golden Gate C2

Keynote
Current Applications and Future Directions for Text Analytics

What is the future for text analytics? Will it continue to be dominated by text mining applications which offer the advantage of automatic processing? Will the Social Media boom go bust or settle into a mature but hardly revolutionary market? Will the market for text analytics continue to grow 20% a year or will it crash or take off? Will there be revolutionary new techniques/applications, or will we see a slower evolutionary growth?

Join text analytics expert, Tom Reamy, in exploring these questions and exploring how organizations are using text analytics today and how they see text analytics in their future.

Join us for a comprehensive view of text analytics applications today and a speculative look at the future of text analytics – and bring your own ideas to the discussion.

Speaker: Tom Reamy, Chief Knowledge Architect, KAPS Group

[ Top of this page ] [ Agenda overview ]


10:00-10:30am
Room: Salon 8 & 9

Exhibits & Morning Coffee Break

[ Top of this page ] [ Agenda overview ]


10:30-11:15am
Room: Golden Gate C2

Track 2: ENTERPRISE TEXT ANALYSIS
Topic: New Directions in Text Analytics

Phonemic-Based Text Analytics: Unlocking the Information in Sound and Narrative

Research has explored how phonemes can be utilized in sentiment detection. Independently of any dictionary meanings, the phonemes in words carry sentiment that can be measured and the timing of delivery also carries meaning. By using a clustering approach similar to a musical chord, it is possible to interpret phonemes in text contextually and have their implied sentiments serve as critical signals for enhanced narrative analysis. As a result, pivotal points in a narrative where the story line shifts are more easily identified. This enables for greater accuracy and discovery of key themes, genres and topics in social media or any form of unstructured text. In this session, you will learn about these new approaches to sentiment and narrative analysis and see examples of how they are applied to a variety of text ranging from social media to speeches to novels.

Speaker: Eric Forst, CMO, Synapsify, Inc.

[ Top of this page ] [ Agenda overview ]



11:20-12:05pm
Room: Golden Gate C1

Track 1: SOCIAL MEDIA / SENTIMENT
Topic: Customer Focus

Case Studies: CFPB, Capital One, Citibank, Bank of America
Flip the Script: Leveraging Social Media Complaints to Improve Customer Experience with Text Analytics

While organizations have historically utilized customer feedback to make business adjustments, the location and depth of this data has shifted drastically in recent years. As social media adoption continues to increase, consumers are turning to outlets, such as Facebook and Twitter, to share positive and negative experiences about the organizations they do business with. In this presentation, Steven Ramirez provides expert insight on leveraging text analytics to unlock valuable customer data from a variety of sources, such as the CFPB’s consumer complaint database and various social media outlets, to identify these customer issues before they spell disaster for businesses.

Speaker: Steven Ramirez, CEO, Beyond the Arc, Inc.

[ Top of this page ] [ Agenda overview ]


11:20-12:05pm
Room: Golden Gate C2

Track 2: ENTERPRISE TEXT ANALYSIS
Topic: New Directions in Text Analytics

Case Study: Keck School of Medicine – University of Southern California
Deep Distillation from Natural Language Text

This presentation examines a next-generation text analytics system that is pioneering the distillation of “expressions” and “contextual sentiment” from natural language text. Attend and explore techniques that combine Natural Language Processing (NLP), Machine Learning and Semantics to achieve significantly higher accuracy in text mining than currently available. This session demonstrates the core text understanding engine, as well as applications powered by it, for unstructured data analytics in domains such as retail, finance and consumer health.

Speaker: Dr. Naveen Ashish, Chief Scientist, Cognie and Associate Professor, Informatics, Keck School of Medicine – University of Southern California

[ Top of this page ] [ Agenda overview ]


12:05-1:15pm
Room: Salon 8 & 9

Lunch

[ Top of this page ] [ Agenda overview ]


1:15-2:00pm
Room: Golden Gate C2

Ask the Experts

A Panel of 4 text analytics experts representing leading vendors, industry watchers, and academics discuss their latest offerings and ideas, but will mainly be available to answer your questions. Bring your thorniest problems!

Moderator: Tom Reamy, Chief Knowledge Architect, KAPS Group, LLC

Speakers:
Toby Conrad, SVP Americas, Smartlogic
Catherine Havasi, CEO, Luminoso
Andrew Fast, Chief Scientist, Elder Research, Inc.

[ Top of this page ] [ Agenda overview ]


2:00-2:15pm
Room: Golden Gate C2

Lightning Round
Expert SystemGnip

[ Top of this page ] [ Agenda overview ]


2:15-3:00pm
Room: Golden Gate C2

Mini-Workshop: Quick Start for Text Analytics

While text analytics solutions offer the hope of beginning to tame the growing information overload, few organizations have the experience to confidently assess what text analytics is, what it can do for their business, or how to select the right tools and right methodology to get the most out of text analytics. Without this experience, organizations can either miss out on a powerful new tool or worse pick the wrong tool and the wrong implementation strategy and waste their time and resources without solving their information overload problems.

This mini-workshop presents a methodology for getting started with text analytics that ensures that organizations will get the most out of text analytics. It consists of three elements. First, an audit that assesses the organization’s content and content structure, the information behaviors and needs, and their people and technology. This produces a text analytics strategy. Second, a text analytics software evaluation process that helps the organization navigate the ever changing vendor landscape. Third, an initial pilot project that creates the foundation for future text analytics applications development, including initial taxonomy development or assessment and training of internal resources.

We will end with a group exercise knowledge audit.

[ Top of this page ] [ Agenda overview ]


3:00-3:30pm
Room: Salon 8 & 9

Exhibits & Afternoon Break

[ Top of this page ] [ Agenda overview ]


3:30-4:15pm
Room: Golden Gate C1

Track 1: SOCIAL MEDIA / SENTIMENT
Topic: TBA

Case Study: 2013 Presidential Inauguration
Revealing Hidden Insights in Social Media Data

In the age of social media, millions of public online interactions are happening every day. These interactions create massive public social data repositories, with each interaction documenting insights into the behaviors and sentiment of consumers. When looking at social data as a whole, it can quickly become overwhelming.

This session examines how public data available on Twitter was used to decipher insights into the 2013 inaugural audience (#inaug2013) and uncover the hidden data within social media. Attend and hear how machine learning algorithms were used to automatically decipher social signals and discover important unknowns within massive text-based social streams. Alexis Johnson, Data Scientist from Ayasdi, shows you how this was done and the findings:

  • Learn how to segment Twitter users into distinct segments based on word usage, meaning similar Twitter users group together based on conversations.
  • Based on word usage, explore hidden themes within conversations automatically.
  • Select a group of tweeters previously identified, create a new topological network to gain a deeper understanding of these specific users. For example, we learned that Spanish speaking tweeters were broken into distinct groups: non-political and political tweeters. This allowed for greater insights into the #inaug 2013 audience – an insight that most analytical tools would have missed.

Speaker: Alexis Johnson, Data Scientist, Ayasdi

[ Top of this page ] [ Agenda overview ]


3:30-4:15pm
Room: Golden Gate C2

Track 2: ENTERPRISE TEXT ANALYSIS
Topic: TBA

Impact of Process on Terminology

Business terminology changes as work items move through their lifecycle. What impacts the terms? How do we manage the complexity of modeling those changes? How can we write effective rules to account for the changes over time or in situ? Internal and external groups can impact how an object is described via communities of practice, schools of thought, marketing/pr practices, negotiated regulatory language and more. This talk discusses the “gotchas” to watch for, the strategies to mitigate their impact, and how to assess if additional modeling work is worth the effort.

Speaker: Christine Connors, Founder & Information Strategist, TriviumRLG

[ Top of this page ] [ Agenda overview ]



4:20-5:05pm
Room: Golden Gate C1

Track 1: SOCIAL MEDIA / SENTIMENT
Topic: Social Media

Amp Your Social Marketing

Social media is a powerful marketing channel, but for most social media marketers, efforts are limited to a single company voice. Being socially active takes time and so does finding the content. Employees want their company to be successful and they like talking about what they do, but finding relevant content can be challenging and most companies don’t have a consistent way for marketers or employees to share content across social channels. Learn how to align, simplify and measure your organization’s social publishing efforts companywide and better understand how social media marketing can drive new and existing business.

Speakers:
Eric Roach, CEO and Co-founder, EveryoneSocial

[ Top of this page ] [ Agenda overview ]


4:20-5:05pm
Room: Golden Gate C2

Track 2: ENTERPRISE TEXT ANALYSIS
Topic: TBA

Case Study: A Federal Government Agency
Optimizing Findability with Text Analytics and Taxonomies

Taxonomy design has become an integral component of most organizations’ information management strategies. Taxonomies are a cost-effective and highly impactful component of successful search, discovery, and findability solutions. This presentation provides a practical approach for taxonomy design and also details a real-world case study from the federal government that demonstrates how Open Source Text Analytics tools and a business taxonomy were leveraged to improve the findability of one of the largest and most complex search applications in the world.

Speaker: Zach Wahl, President and Chief Executive Officer, Enterprise Knowledge

[ Top of this page ] [ Agenda overview ]

 

Free Book Chapter