You are on page 1of 31

Beyond

Sen)ment Hype: Conversa)on Context for Accurate Discovery


Hadley Reynolds
NextEra Research

Agenda
Where we are now market drivers & technology dynamics The Sen)ment Bubble considered Dieren)a)ng levels of analysis Prac)cal dimensions of analysis and examples Discussion

Market Drivers for Sen)ment Analysis

Market Drivers for Sen)ment Analysis

Market Drivers for Sen)ment Analysis

Market Drivers for Sen)ment Analysis

Market Drivers for Sen)ment Analysis

Addi$onal Web 2.0 Content: Blogs Discussion Forums Amazon (Yelp, Trip Advisor etc.) Reviews User Generated RaAngs Data Like Google+ And more, much more

Sen)ment Technology Providers


45 40 35 30 25 20 15 10 5 0 2003 2004 2005 2006 2007 2008 2009 2010 2011
Corpora Software

Where Does Sen)ment Belong?

Early Social Monitoring

Nave Sen)ment

Credibility: comes from accuracy & insight Ambiguity: is the enemy of accuracy

Challenges for Sen)ment Analysis


Level of analysis Timeframes for analysis Rela)ve sophis)ca)on of analysis

Level of Analysis
Corpus (Do the bloggers like us?) Document (Does this author like us?)

Document Sen)ment Math


Posi)ve document = 4 points or above Nega)ve document = -2 points or below Neutral document = -2 through +3
good great o.k.

Term good great o.k. disappointed Total:

Value

Score

2 3 1 -4

2 3 1 -4 +2

disappointed

Neutral Document

Document Sen)ment Math


Product A good great o.k. Product B ok good ok

Posi)ve document = 4 points or above Nega)ve document = -2 points or below Neutral document = -2 through +3 Term Product A good Product A great Product A o.k. Product A disappointed Product B good
Value Score

2 3 1 -4 1 1 -4

2 3 1 -4 1 2 -8

disappointed disappointed disappointed

Product B ok Product B disappointed Total:

Nega)ve Document

-3

Level of Analysis
Corpus (Do the bloggers like us?) Document (Does this author like us?) Sentence (What is this persons comment?) En)ty/A`ribute (What is it about us that she likes or doesnt like?)

En)ty-level Analysis
Sources

Person

Opinion Target En)ty

(Feature) (Prole) Person (Emo)on) Opinion (Feature) Target En)ty (Feature) (Social Network)

Timeframes of Analysis
Retrospec)ve analy)cs/business intelligence Predic)ve analy)cs quality issues, future performance Trend emergence Real-)me customer interac)ons, social interac)ons/engagements

Sophis)ca)on of Analysis
Keyword-based sen)ment techniques
Sen)ment terms: elusive, ambiguous, in ux Sen)ment lexicons: incomplete, non-specic, inexible Unable to understand context surrounding an expression or the people contribu)ng Unable to understand connec)ons among related en))es and a`ributes and people Unable to gauge quality of source materials

Sophis)ca)on of Analysis
Seman)c-based sen)ment techniques
Sen)ment terms >> incorporate related expressions, fuzzy logic - NLP Sen)ment lexicons >> domain ontologies (available or buildable) provide analy)cal context Able to understand context surrounding an expression or the people contribu)ng - machine learning & other techniques Able to understand connec)ons among related en))es and a`ributes and people - triples, event extrac)on

Dimensions of Analysis
Ontologies around opinion objects Iden)ca)on and qualica)on of en))es & a`ributes & rela)onships Emo)onal content of expression(s) Quality gauge of sources Proles of individual commenters Roles/interac)ons/sociology of commenters and their alia)ons Timeframe for expressions and responses

Beyond +/-: Ontology-based analy)cs


Same Ontology breakdown Same Scale: Expressed Opinions

Higher values for cardiovascular diseases with Avas)n

Source: BuzzStory

Opinion::Emo)on

Source: BuzzStory

Quality of Content Sources


topix.com cancergrace.org

Quality: 4.48
"I know of one method that would be really scary and graphic that would work towards gepng people to stop pollu)ng my sea breeze environment. What I wonna know is they keep pupng down smokers and blaming us for evrything.

Quality: 16.78
"As shown above, a total of 362 pa)ents who hadn't progressed aser rst line chemo/Avas)n were randomized to either of the two maintenance therapy arms, and the combina)on arm showed a signicantly longer progression-free survival (PFS) coun)ng from the beginning of all treatment, at 10.

Alia)on Network Map of Alia)ons of People & Topics


Supplements Tobacco Addic)on Prostate Cancer Breast Cancer

Co-Morbidi)es

Thyroid Disease Biomarkers Targeted Therapies Lung Cancer Chemotherapy

H&N Cancer

Source: BuzzStory

Sociology of Alia)ons & Topic Groupings


Tobacco Addic)on Co-Morbidi)es

Other Types of Cancer

Supplements

Misc. Side-Eects

Biomarkers

Misc. Side-Eects

Source: BuzzStory

Where Does Sen)ment Belong?

Contextual Analy)cs

Keyword technology

Challenges Remain
The service at Reynards is, in general, friendly and loose. Though they couldnt nd a reserva)on for four one Friday night, they compensated with so much warmth and comped wine that all was forgiven. In some ways, Reynards oers what one wishes a dining experience in Manha`an would be: kindness instead of aptude, inoensive prices, glorious food, and aesthe)c varietythe clientele is split roughly in half between the stylish and the schlumpy.
The New Yorker, September 24, 2012

Resources
Bing Liu, Sen$ment Analysis and Opinion Mining, Morgan & Claypool, 2012 Bo Pang and Lillian Lee, Opinion Mining and Sen$ment Analysis, (Founda$ons and Trends in Informa$on Retrieval), Now Publishers, 2008 Sen)ment Analysis Symposium, San Francisco, CA, October 30, 2012

Ques)ons?

hadleyr@nexteraresearch.com

You might also like