The next important piece (and, I think, one of the most critical) is the ability to lend structure to that unstructured text. To be able to “read” documents in many different languages and pull out the who’s, when’s, what’s, and where’s contained within result sets – things like Who: Person, Company, Organization, Financial Index, Social Security Number When: Date, Day, Holiday, Month, Year, Time, Time Period Where: Address, City, State, Country, Place (Region, Political, Geocoordinates) Other: Product Name, Vehicle (Make, Model, Color, VIN, License Plate), Currency, Measure, Internet Address, Phone Number Concepts (Global piracy, unstructured data…) And custom “entities” such as company specific SKYs, code names, etc. It also should be able to pull out events (such as company A merged with company B) and relationships (such as Bob Smith is the VP of Sales at Company C or Kelly knows Bob). It needs to be able to ISO standardize and normalize this information to better enable data mining as well.

Where a user can then access information derived from lithium-powered forums, together with FAQs, service manuals, and other infromation

Where a user can then access information derived from lithium-powered forums, together with FAQs, service manuals, and other infromation

Here’s an example of an intent-to-purchase being routed to the sales department. Collaboration features enable multiple people to work to solve the customer’s issues before sending a resolution message.

. Here we see why the customer bought the product (when it was mentioned in the review). You can see that the majority of people are writing about their LCD TV, and that the majority of people bought something for their bedroom. Now let’s focus in specifically on purchasing reasons for laptops… <click on Laptop>.

For example, we see here that the majority of laptops were purchased for the reviewer’s wife. We could create targeted marketing campaigns that leverage our customer knowledge to target those customers that are married that would to drive them to purchase a laptop for their wives. Here you can see one of the great things that you can do with Attensity – discovery. For example, it’s unlikely that you would have put “bought for wife” as a category right out of your head. But attensity extracted this information automatically and revealed this deep insight. <switch to text preview and click on actionable details/Bought For Wife>

Note that in all cases the laptop was purchased for a wife – you can see here some of the difficult constructs that Attensity can handle. for example in several cases, only in the previous sentence was a “laptop” mentioned. The next sentence says “I bought this for my wife”. Attensity uses what we call anaphora resolution to resolve a pronoun back to the original referent noun. Let’s focus in on the second item, “Bought this for my wife for her birthday”…

Not only can we identify who the purchase was for, we can identify why it was purchased, in this case for her birthday. Notice in the first four triples, we see not only that she had a birthday, and that it was bought for her, but that it was bought for her birthday. It is this level of automated detail and accuracy that makes Attensity unique! By understanding purchasing motives, as found in the review text, you can create more focused, more targeted marketing campaigns which will drive additional revenue at a lower cost.

Thanks for joining us for these past few minutes. We’ve shown you just a glimpse into how (read slide)

LEVERAGING LITHIUM COMMUNITIES AND OTHER CUSTOMER CONVERSATIONS THROUGH LARA Catherine van Zuylen VP of Product Marketing Attensity Americas

A Few Words About Attensity Attensity: Over 20 years experience understanding customer conversations in text; 6 patents in natural language processing Suite of applications for social media monitoring, Voice of the Customer Analysis, and Self-Service/Agent Service Over 500 customers worldwide

“Customer Information” is changing and growing exponentially Twitter hit the 10 billion tweet mark last week : over 20% are about products and services Over 247 billion emails are sent every day Millions of customer interaction records in a typical large company.

To effectively harness these “customer conversations”, you need a program to comprehensively Listen across customer conversation channels Analyze accurately and efficiently Relate this information to other information Act on the information We call this the LARA methodology

LARA Methodology: Listen, Analyze, Relate, Act Are you listening where your customers are talking? Are your “social media” listening efforts isolated from your “CRM” listening efforts and separate from your “survey” listening? Are you monitoring those internal customer communities that you’ve set up with Lithium? Text Analysis can help bridge these gaps.

Text Analysis is not Search “Search” is for finding relevant or recent documents that contain a term of interest

But it’s hard with search to get the “big picture” 7 What do people think about my company? What problems are they having? Who is thinking of switching? What do they like about me vs. the competition? What new ideas do they have?

“Search” starts with you feeding a system words to look for. “Text Analysis” starts with the data itself and lets it tell a story Documents Dynamic Text Profiling ? Entities, sentiments, events and relationships, intent, etc XML or other “tags”

Text Analysis starts the same way some search engines do… Automatic Language and Character Encoding Identification Identify paragraphs and sentences within text Word Segmentation (Tokenization) and De-Compounding Part-of-Speech Tagging Stemming Noun-Phrase Identification

Then continues with Entity Extraction… Who: People, Person Position, Social Security Numbers What: Companies, Organizations, Financial Indexes, Products (software, weapons, vehicles, etc…) When: Dates, Days, Holidays, Months, Years, Times, Time Periods Where: Addresses, Cities, States, Countries, Facilities (stadiums, plants), Internet Addresses, Phone Numbers How Much: Currencies, Measures Concepts (i.e. Global piracy, unstructured data…) Can be pattern-based – tell the system that a “Prop-Noun followed by Smith” is probably a person Or machine learning – feed it a million proper names and let it deduce names from those examples…

Practical Text Analysis in Action Let’s say that I am a major retailer, and someone posted a review in my Lithium-powered forum that starts out I bought this Gucci scarf for my mom in your Santana Row store last week. Entities (brands, people, locations, times, products…)

To “connect the dots” in data, you also need to extract noun-verb relationships, sentiment… I bought this Gucci scarf for my mom in your Santana Row store last week. I really like the pattern, but I don’t like how it itches. Entities (brands, people, locations, times, products…) Events and relationships: action and purchasing reason Sentiment (extreme positive, positive, negative, extreme negative)

To “connect the dots” in data, you also need to extract suggestions, intent… I bought this Gucci scarf for my mom in your Santana Row store last week. I really like the pattern, but I don’t like how it itches. I wish this scarf came in cotton. If Gucci made more cotton scarves, I would buy them all. Entities (brands, people, locations, times, products…) Events and relationships (I : buy : this Gucci scarf | I : buy : for mom) Sentiment (extreme positive, positive, negative, extreme negative) Suggestions (I : wish : this scarf came in cotton) Intent (to purchase, to leave) (If Gucci made more cotton scarves, I would buy them.)

How do you do this? You parse sentences like a human…and extract triples…

…and voices (intent, recurrence, etc) Question [?] voice: How can I get free shipping with future orders?   Condition [if/then] voice:. I would shop more frequently if you offered free shipping.   Intent [intent] voice: I plan to place an order today.   Negation [not] negates the meaning of the verb: You did not have the size I was looking for in stock  

…and voices (intent, recurrence, etc) Question [?] voice: How can I get free shipping with future orders?   Condition [if/then] voice:. I would shop more frequently if you offered free shipping.   Intent [intent] voice: I plan to place an order today.   Negation [not] negates the meaning of the verb: You did not have the size I was looking for in stock   Augment [more] voice: The staff were incredibly professional   Recurrence [again] voice: I had to enter my information several times for the order to process   Indefinite voice representing suggestions or requests. You should sell wedding dresses, too!

LARA Methodology: Listen, Analyze, Relate, Act Once you’ve done text analysis, you can relate the text to structured information… 01/24/2010 By errodd from San Jose, CA I bought this Gucci scarf for my mom in your Santana Row store last week. I really like the pattern, but I don’t like how it itches. I wish this scarf came in cotton. If Gucci made more cotton scarves, I would buy them all. Can help you answer questions like What were the top concerns of people who rated this product a “4”?

LARA Methodology: Listen, Analyze, Relate, Act: What Can You Do with Text Analysis? The output from text analysis can be exported as XML… It can also be used directly in applications that Seek out and deliver information to those who need it Route and respond to communications Mine and report on information

“Seek Out” information for a self-service knowledgebase Problem Solution Manufacturer: Apple Product: Macbook, Projector, Monitor Component: Adapter cord, Mini-DVI, VGA Action: Do a presentation, connect

Route and respond to all customer communications Refund policy? Email Issue posting that has not been answered in 8 hours by the community Threatening to sue posting “refund policy” email response auto-generated Automatically routed as a mobile alert to legal for review Responses can be reviewed by agent before sending Read text and extract knowledge about what the document is saying People Places Events Topics Sentiment … Routed to Customer Service for Follow-up and Resolution

For example, here you can see a unanswered “cry for help” being routed to a customer service agent for response. They can see the message, and can click to respond quickly.

Here’s an example of an intent-to-purchase being routed to the sales department. Collaboration features enable multiple people to work to solve the customer’s issues before sending a resolution message.

Within the system, you can also respond directly to the poster on the forum with a simple “click to post” The system can suggest a response based on your corporate knowledgebase or you can create a response on your own.

Mine and report on sentiments, complaints, compliments, reasons for purchase, and “intentional” behavior across all customer conversations

For example, let’s take a look at this “Purchasing Motives” report, which draws from information in a Lithium-powered forum

Focusing down on laptops, you can see that the majority of laptops purchased were purchased for the poster’s wife

Together, Attensity and Lithium Enable You to More Effectively

For more information sales@attensity.com Follow Attensity on Twitter @attensity Follow Catherine van Zuylen on Twitter @catevz sales@lithium.com Follow us on twitter @lithium