Sep 24, 2012 (11:09 AM EDT)
Where Sentiment Analysis Heads Next
Read the Original Article at InformationWeek
Whether in conversation or posted online or to our social networks, subjectivity and sentiment add richness to human communications. Captured electronically, customer sentiment--expressions that go beyond facts and that convey mood, opinion, and emotion--carries immense business value.
We're talking about the voice of the customer, and of the prospect, patient, voter, and opinion leader.
Listening--for brand mentions, complaints, and concerns--is the first element of any credible social engagement program. Businesses that listen can uncover sales opportunities, measure satisfaction, gauge reactions to marketing campaigns and message themes, uncover root causes behind events, and detect and respond to reputation and competitive threats. That's why we have monitoring and analytics solutions--the best of which apply text and sentiment analysis technology--targeting online and social media as well as enterprise feedback in surveys, e-mail, and contact center notes. These solutions are aimed at discovering business value in complex, expressive, and sometimes-confusing human language.
My aim here is to explore modalities, or how information technology helps us get at affect and attitudinal information. (These points will be covered in depth at a social media analysis and engagement conference that I organize: the Sentiment Analysis Symposium, October 30 in San Francisco.)
[ Is your organization keeping up with all the ways customers connect? Read more at How To Scale Social Business For The Real World. ]
I'll start with a key definition: Sentiment analysis systematically rates human affective states according to positive or negative polarity or a neutral or mixed value, or according to mood, emotion, or feelings (angry, happy, sad, proud, disappointed, etc.) and to use sentiment data for business purposes.
Then we'll explore two questions: What types of sentiment analysis are there? And in what directions is sentiment analysis evolving?
Six Types of Sentiment Analysis
"What type of person is she?" The question has many answers.
Each of us has many types, according to our demographic category (e.g., sex, age, race, income), personality, interests, occupation, and so on. We are many types simultaneously. Any question that takes the form, "What types of x are there?" has many answers, and when x = Sentiment Analysis, there's no exception.
Of course, we're most interested in the most important types of sentiment analysis. Here they are, as I see them:
Coarse-grained to fine-grained. Some analyses discern sentiment at a corpus or data-space level (e.g., for a set of reviews or survey responses); others score particular documents or messages; and others resolve sentiment at an entity (e.g., person, place, or company), topic, or concept level. Coarse-grained analysis is fine for some business applications, but others need fine-grained.
Individual versus aggregate. Analyses might look for individual cases ("mentions") or for aggregates over populations or sources or trends over time. If you're managing customer support, for example, you need to get at each mention, but if you're studying market pulse you're looking for the big picture.
Metric. Analyses may rate sentiment on an absolute scale, or they may look for relative/comparative sentiment--"I don't much care for sports, but I do prefer basketball to ice hockey"--and measure variation, intensity, and change.
I believe these three factors are underappreciated, underutilized measures. Too many tools rate a review with 8 positive points where 7 negative is mildly positive. But the high degree of variation should flag the review as interesting, even more so if there are both strong positives and strong negatives--that is, intense versus mild opinions. Sentiment change is always notable. It invites the question: What triggered the change?
Focused or integrated. Do you need to go beyond simple figures to get at breakdowns, root causes, and predictions?
Integrated analyses seek to link sentiment to psychological profile, behaviors, demographic characteristics, transactions, events, and/or other data.
How It's Done. Here are five different functional sentiment analysis techniques:
1) Analysis by trained analysts.
2) Crowd-sourced analysis by untrained humans.
3) Automated analysis of information extracted from unstructured sources such as text, audio, images, and video--for instance, for text via natural-language processing (NLP).
4) Analysis of categorical poll or survey questions--for example, "Rate your hotel stay on a scale of 1 to 5," or the equivalent star ratings.
5) Inference of sentiment from numerical statistics--for instance, commercial inventories, consumer spending, investment levels, etc. Automated NLP may apply linguistic, statistical, and/or machine-learning techniques.
ROI. There's sentiment analysis that delivers business value, and there's eye candy. We see lots of social media analytics and BI dashboards that convey sentiment via pie charts, trend lines, color-coded word clouds, and other graphics. With most of these, I've concluded, you face a decision gap: They give you information (sometimes accurate, sometimes not), but they don't convey meaning or suggest how you can use the information to improve your business decision-making.
The type of sentiment analysis that will work for you is analysis that is aligned to your business goals. Work back from your goals to understand the type of insights that will best help you make better business decisions. Decide what data sources you need to tap and which analytical techniques and level of granularity are appropriate. Determine exactly what you're going to measure and how you're going to link sentiment to the wealth of other types of data available. Then you'll be positioned for return on investment--the only type of sentiment analysis that really matters: sentiment analysis that delivers business value.
Let's conclude with a look ahead. Sentiment analysis is evolving in the following important directions:
--As an industry. Adoption of sentiment analysis is growing in a spectrum of business domains and applications. The anti-automation backlash continues, but it should fade as sentiment analysis providers and users move toward semantically infused analysis with feature-level, business-need-aligned sentiment resolution and away from simplistic, keyword-based solutions.
--Considering sources. Detection and exploitation of emotion in speech and images (such as facial and body language) implied by video-captured behaviors will increasingly come into play, including actively in meeting commerce and security needs. Availability of solutions for smaller-market languages will remain demand-driven.
--A focus on intent. We seek to understand not just how people feel, but what their feelings, linked to data from a variety of relevant associated sources, say about plans.
--Predictive modeling. Sentiment analysis becomes fuel for efforts to shape opinion, attitude, and emotion.
The end result is sentiment analysis as a contributor to sense-making, to intelligent automation that enables machines to understand and act on the spectrum of signals present in the human world.
Seth Grimes is an analytics industry observer--an analyst, consultant, and writer--who helps organizations find business value in enterprise data and online information. Seth consults via Alta Plana Corporation, works as an industry analyst, organizes the Sentiment Analysis Symposium, and tweets at @sethgrimes.
Social media make the customer more powerful than ever. Here's how to listen and react. Also in the new, all-digital The Customer Really Comes First issue of The BrainYard: The right tools can help smooth over the rough edges in your social business architecture. (Free registration required.)