Mining for What’s Missing: How to Find What’s Not in the Speech Application’s Vocabulary AMY NEUSTEIN, Ph.D. LINGUISTIC TECNOLOGY SYSTEMS

Slides:



Advertisements
Similar presentations
Testing Relational Database
Advertisements

Access 2007 ® Use Databases How can Microsoft Access 2007 help you structure your database?
AMEP Assessment Task Bank Professional Development Kit Question Items
Teaching grammar better! Hugh Dellar The University of Westminster Heinle Cengage.
Word Lesson 8 Increasing Efficiency Using Word
MIS 5241 SOFTWARE. MIS 5242 Agenda The Stored Program Concept Software as Control Software as Simulation.
Automating Tasks With Macros
Visual Cognition II Object Perception. Theories of Object Recognition Template matching models Feature matching Models Recognition-by-components Configural.
Downloading and Installing AutoCAD Architecture 2015 This is a 4 step process 1.Register with the Autodesk Student Community 2.Downloading the software.
1 How To Annotate Interactions Using Dialog Function Units (Part 1) by Michal Novemsky (with the help of Becky Passonneau & Eddie Kang) CCLS, Columbia.
ERIN STALBERG NCSU LIBRARIES SEPTEMBER 16, 2009 Cool Tools – More Connexion.
 Bachelors degree: 75% of teachers have the bachelors degree  Masters degree: 25% of teachers have the masters degree.
Statistical Natural Language Processing. What is NLP?  Natural Language Processing (NLP), or Computational Linguistics, is concerned with theoretical.
Allied Health Assisting
Dragon Naturally Speaking Tutorial What is Dragon Naturally Speaking? Dragon is a dictation software, students can dictate a paper rather than type it.
ACOS 2010 Standards of Mathematical Practice
Jude Carroll, author of Tools for Teaching in an Educationally Mobile World (Routledge 2015) Supporting teaching across cultures: the role of good practice.
GRAMMAR IN SPEECH AND WRITING. A12.1 Variety in English ❏❏ between different dialects of English, for example, British and American forms e.g. I have.
1 A Practical Rollout & Tuning Strategy Phil Shinn 08/06.
Microsoft Office Word 2013 Expert Microsoft Office Word 2013 Expert Courseware # 3251 Lesson 5: Setting Up Global Accessibility.
Lesson 6 Part 1. 2 When would the use of a template save time and be more productive? In other literature a “template” may also be referred to as a “boiler.
Challenges in Information Retrieval and Language Modeling Michael Shepherd Dalhousie University Halifax, NS Canada.
Second Annual Research Symposium of the Human Language Technology Research Institute Sequence Package Analysis: A New Natural Language Intelligence Method.
Example Usability Problems for Diagnosis with the UAF.
Designing a Graphical User Interface (GUI) Krisana Chinnasarn, Ph.D. January 2007.
Communicative Resources. How Do We Communicate? Conversation involves more than language – Gestures, facial expressions, tone of voice, … – Face-to-face.
THE TBL FRAMEWORK: LAGUAGE FOCUS Willis, J. (1996) ByJulietaEdayFabiola.
Beyond Call Recording: Speech Improves Quality Assurance Larry Mark Chief Technology Officer SER Solutions, Inc.
Turning Audio Search and Speech Analytics into Business Intelligence.
Speech Analytics ROI: Uncovering Key Business Intelligence Can Save Revenue From Dropping off the Bottom Line.
Some initial habits include: Writing down everything the instructors says Poor organizational structure within notes (hierarchy of ideas) Failing to recognize.
A pattern for a listening session Pre-listeningWhile-listeningPost-listening.
Teaching Productive Skills Which ones are they? Writing… and… Speaking They have similarities and Differences.
Reviewing Recent ICSE Proceedings For:.  Defining and Continuous Checking of Structural Program Dependencies  Automatic Inference of Structural Changes.
SEQUENCE PACKAGE ANALYSIS: A NEW WAY TO UNDERSTAND NATURAL LANGUAGE DATA ACROSS DIFFERENT LANGUAGES AND DIALECTS AMY NEUSTEIN, Ph.D. LINGUISTIC TECNOLOGY.
Computing Fundamentals Module Lesson 3 — Changing Settings and Customizing the Desktop Computer Literacy BASICS.
Socratic Seminars EXPECTATIONS FOR A SUCCESSFUL DISCUSSION.
SEQUENCE PACKAGE ANALYSIS: A New Natural Language Understanding Method for Performing Data Mining of Help-Line Calls and Doctor- Patient Interviews AMY.
© 2008 The McGraw-Hill Companies, Inc. All rights reserved. M I C R O S O F T ® Refining Original Illustrations Lesson 9.
Lesson Understand templates 2 Create a new document from a template 3 Work with template elements 4 Create a custom template 5 Use a custom template.
12.1 CSC 123 Systems Analysis & Design Part IV: The Essentials of Design Chapter 12 Designing Effective Input.
Comparing and Ranking Documents Once our search engine has retrieved a set of documents, we may want to Rank them by relevance –Which are the best fit.
Copyright 2006 by Timothy J. McGuire, Ph.D. 1 MIPS Assembly Language CS 333 Sam Houston State University Dr. Tim McGuire.
CHAPTER 19 Communication Skills.
Sight Word List.
MICROSOFT WORD 2010 Lesson 6: Word Templates. The goal of this lesson is for the students to successfully create and work with templates. The student.
Computing Fundamentals Module Lesson 7 — The Windows Operating System Computer Literacy BASICS.
© 2013 by Larson Technical Services
YALE LAW SCHOOL POLICY SCIENCES CENTER ANNUAL INSTITUTE Using a New Method of Natural Language Intelligence for Performing Wiretap Analysis Amy Neustein,
Sight Words.
1 Core English 1 Listening Task – p 158 Rhetorical Function Questions.
Attention Boosters in ESP Lectures. General Problem language teaching requires the student’s active participation in the process to turn out successful.
Copyright 2006 by Timothy J. McGuire, Ph.D. 1 MIPS Assembly Language CS 333 Sam Houston State University Dr. Tim McGuire.
The Non Fictional “Hal”? Rather than make humans conform to computer-speak, design computers to understand conversational dialog. KeyWordsSequence Packages.
Sequence Package Analysis A New Data Mining Tool to Speed Up Wiretap Analysis Amy Neustein, Ph.D. Linguistic Technology Systems
Year R Stay and Play Talk. Why?  Communication is the number one skill. Without it, children will struggle to make friends, learn and enjoy life.
CELDT PRACTICE Speaking Version B.
LEARNING UNIT 7 (Week 11) Making A Business Telephone Call ENGLISH FOR PROFESSIONAL COMMUNICATION.
CELDT: Speaking Practice MVSD Grades K-1 Adapted from LAUSD CELDT Resources.
CELDT PRACTICE Speaking Version A.
GRADE 5 Copyright © 2015 by Write Score LLC. What to Expect when Finding Evidence in Sources: Today, we are going to work on how to find, sort, and select.
ENGLISH FOR PROFESSIONAL COMMUNICATION
Microsoft Word 2016 Lesson 6 Part 1.
DATA INPUT AND OUTPUT.
Adding Assignments and Learning Units to Your TSS Course
Requirements Reference: Software Engineering, by Ian Sommerville, 6th edition, Chapters 5, 6, & 8.
A New Conversational Query Language (C-QL) For The “Emotionally Intelligent” Smartphone Amy Neustein, Ph.D.
Building and Integrating a Chatbot in 30 minutes
Communicative Resources
Education & AI High level discussion points relating ‘techniques / tools’ to possible ‘projects’
Presentation transcript:

Mining for What’s Missing: How to Find What’s Not in the Speech Application’s Vocabulary AMY NEUSTEIN, Ph.D. LINGUISTIC TECNOLOGY SYSTEMS SpeechTEK 2004

First Problem: Critical business intelligence data is lost in a sea of recorded calls when callers use words outside of the application’s vocabulary

Second Problem: Early warning signs of caller frustration are hard to detect when callers do not use expected “keywords” from the application’s vocabulary to express frustration

Third Problem: To build a Statistical Language Model to accommodate all the ways users might express themselves would require a very large data corpus that is costly to assemble; and still there would be no guarantee that an accurate word match would be found.

THE SOLUTION: SEQUENCE PACKAGE ANALYSIS A new natural language intelligence method that has been successfully peer reviewed; and cited by other researchers as a data mining method for call center quality monitoring.

METHODOLOGY SPA draws mainly from the field of conversation analysis: the study of the orderly properties of interactive dialog that revolve around the turn-taking process; and other sequentially based features that are part of that process such as spacing between turns and overlap of turns

How Does Sequence Package Analysis (SPA) Work? SPA parses NL dialog to locate a series of related turns, discretely packaged as a sequence of conversational interaction. SPA locates generic sequence packages, rather than isolated key words, because speakers are more likely to vary in their choice of words than in their basic conversational sequence patterns.

SPA provides a “filter” for the front end of a speech recognizer, using generic templates that can be deployed in many different applications and languages. A SPA “add on” layer can be used with conventional vector-based n-gram language models, which hold spaces and determine “global weighting” of specific lexical items. WHERE DOES SPA FIT ON THE SPEECH RECOGNIZER?

MINING HELP-LINE CALLS Using SPA to caption the text of a help - line call to capture signs of caller frustration SPA mining tools are based on the detection of conversational sequence patterns rather than solely on word spotting (“get me a supervisor!”) or changes in prosody (e.g., increased pitch) While speakers can vary widely in their choice of words or in stress patterns, conversational sequence patterns are more consistent across a wide spectrum of callers

Australian Help-Line Desk Caller: “I’ve installed Office 97 and…I was a bit stupid. I went into uninstall and um pulled off a whole stack of items off the uninstall and it was a very silly thing to do so now when I start up my computer I get a screen um which say um a black- a black and white screen which says never delete this item. It’s a message screen and every time I start up it comes up……[deleted text]……... Caller: “I’m wondering if I reinstall will I wipe out [my documents]” Agent: “Okay, well look I could certainly have a technician look at the problem for you; we do charge for are you aware of that?” Caller: “I’m just asking a question - I’m just wondering whether or not I should uninstall Microsoft Word?”

Using SPA to Find CONVERSATIONAL SEQUENCE PATTERNS in this Dialog Sample Step One: Locate the pre-question phrases of reports of troubles and requests for assistance: “I’m wondering if” “I’m just asking a question” “I’m just wondering whether or not” Step Two: Quantify the number of times and the proximity of such pre-question phrases. Step Three: Determine if they escalate or, in the alternative, diminish?

ANALYSIS The caller to the Australian help-line began her complaint as a long winded narrative, but with the noticeable absence of a request for help. The caller later produced pre-question phrases when she made her request for help However, these phrases began to escalate (by being combined with one another) just at the point where she began to show signs of frustration: “I’m just asking a question - I’m just wondering whether or not I should uninstall Microsoft Word?”

CODA Conventional data mining programs would have“missed” these signs of caller frustration in that they try to locate keywords and phrases: “get me a supervisor” “I’m frustrated because I’m really not getting answers to my questions.” SPA offers as an add on layer to mining programs in order to locate what is missing