Monday, November 22, 2010

Sentiment Analysis Taxonomy

Sentiment analysis, or Customer Sentiment Analysis, has become an important component for monitoring attitudes or feelings about your business, products, or reputation on the Internet. With social media vehicles such as Twitter, Facebook, product review sites, comment sections, and more, your company or product could be mentioned all over the web. It's important to understand how people feel about your company, and take action to respond if necessary.

WAND has created a taxonomy specifically for this purpose. Our new customer sentiment taxonomy has hundreds of terms including, verbs, adverbs, superlatives, adjectives, and more that all correspond to an intensity of feeling from Extremely dissatisfied, Dissatisfied, Neutral, Satisfied, and Extremely Satisfied.

On twitter, a user might mention "I went to XYZ restaurant and the service was incredible". Using a taxonomy approach, one could understand that "incredible" indicates extreme satisfaction with the restaurant. Or, imagine a product review that says "When I received ABC product, the case was cracked". This would indicate that the product was in disrepair when it was received.

It's easy to understand and comprehend the meaning of the above examples by just reading them. However, with a huge volume of social media and other commentary on the Internet, a company cannot manage to read everything posted about it or its products/services. Leveraging a taxonomy that is designed to identify different levels of sentiment can be a valuable tool for any company to keep tabs on what people are saying about it online.

Sentiment taxonomies are not just valuable in monitoring online and text sentiment, however. They can be equally powerful in helping to analyze vasts amount of customer service or other call center communication. Transcriptions of phone calls can be run against a taxonomy to tag phone calls with levels of satisfaction. This allows a manager to identify an overall level of satisfaction experienced by customer the call-reps are speaking with, but also to identify specific calls that should be listened to and perhaps followed up with.

A recorded customer service call could conceivably last 5-10 minutes. It would be inefficient to listen to the entire call to find the exact point where a customer became upset. Instead, if the customer said "Your company's product is terrible", the exact point in the call where this took place could be tagged by the sentiment taxonomy and a manager could start listening at that point in the call.

For training purposes, a manager could identify all calls where the words "not helpful" were spoken by the customer. The manager could listen to these calls and determine why the representative was not being helpful and identify possible areas of improvement.

Call transcripts or social media could also be monitored and tagged with other taxonomies, such as product names, product features, or other elements to provide the ability to drill through the monitored data from several dimensions.

Please reach out if you'd like to learn more about sentiment analysis taxonomies or how the solutions described above can be implemented.

Thursday, November 18, 2010

Five ways to use taxonomy

I've worked in the taxonomy space for quite a while now, but I still always speak to people who are just learning about taxonomies for the first time. Often times they have a specific project that they are working on that they know a taxonomy is needed. Sometimes, they have a challenge they are not sure how to solve, but think taxonomy may help. Other times, it's some poor person who may have asked me what I do for a living.

For any of these people, here is a list of things that taxonomies can be used for. I'll continue to add to this list and appreciate any input from others on items I may have missed.

1) Directory category structures. This could be a B2B directory, online yellow pages, general web page directory, or more. When somebody is searching for something in a directory, they need to be able to search by a category. Imagine looking through a set of 14 million businesses in the United States and not being able to narrow it down to a list of just the Plumbers. Taxonomy organizes big sets of information in directories so that they can be easily found. The first prominent Internet search engine, Yahoo!, was organized around a directory of websites, where each web site was put into a category that was part of a hierarchy, i.e.; a taxonomy.

2) Enterprise Search. Corporations may have hundreds of thousands or millions of documents in a shared folder, enterprise content repository, or on their intranet. This represents a huge amount of unstructured information - incredibly valuable data but very difficult to wade through. In a physical environment, a company would have put all of this information in filing cabinets with labeled folders. In the digital world, things may not be so organized. People expect to rely on a search engine to help them sort through all this information, but a keyword search may bring back thousands of results. Taxonomy can help filter and drill through these thousands of results. A well designed vocabulary covering terminology that is important to the enterprise can be tagged to the document set. This provides a valuable tool for a user to find the exact document he or she is looking for. In this case, a taxonomy, or corporate vocabulary, enables sophisticated faceted search through corporate documents.

3) Search Engine Optimization. Taxonomy can be used in SEO in a couple of ways. First, it is valuable to organize pages on a website according to a directory structure. Search engine spiders like this and it helps them appropriately index a web site. So, a URL would be structured something like Second, developing keywords and meta-data for any given website can benefit from a taxonomy that can suggest these keywords. This may include brands, other related terms, narrower terms, or term variations.

4) E-Commerce Product Search. Online catalogs need to be organized in a way that makes sense to the shopper. The shopper should be able to quickly get to the product they are interested in so they can make a transaction. The longer a shopper is browsing around a site looking for the right product, the more likely the shopper is to abandon the site entirely. Taxonomy helps this in two ways. First, the category tree itself is a taxonomy. On nearly every e-commerce website, the user is presented with 10-15 (sometimes more or less than this range) top level categories or departments. Clicking on any of these brings the user down the tree to more specific categories. An example may be Office Supplies>Envelopes>Padded Envelopes or Computer Systems>Laptop Computers>Netbooks. This tree needs to be straightforward and easy to navigate. The second place taxonomy is valuable is for providing further attribute drill down at a category level. For example, an e-commerce site may have 100 products in their category laptop computers. A shopper may be interested in a laptop computer that has a 15" screen, a built-in web-cam, and a hard drive larger than 500 GB. A taxonomy can provide these attributes and others as options for the tree category "Laptop Computers". So, instead of browsing through all 100 laptop computers for sale, the shopper can simply select the appropriate attributes and narrow down the results set to the 5 that meet his or her needs. Anytime you see a category tree or product filtering options on an e-commerce site, you are looking at a taxonomy.

5) Improving keyword search. Let's use another e-commerce example. Imagine you are doing a keyword search on an electronics website for "Laptop Computers". This may bring back 50 results, let's say. However, the site also has another 50 products that are labeled as "Notebook Computers". Those results would not be returned so, as a user, you would be presented with an incomplete results set. By making "Laptop Computers" a synonym to "Notebook Computers", a search for either will return the complete results set. That is taxonomy in action. Imagine another example of an online yellow pages search. There are 10 furniture stores within 10 miles of my house. Each of these is in the "Furniture Retailer" category on an online yellow pages website. If i do a search for "Beds" on the local search site, if there is not taxonomy, then that keyword search will not match to the category name, and unless one of the furniture stores has the word "beds" in its name, i will not get any results. Taxonomy can fix this problem. By associating a set of narrower terms from a product and service taxonomy to the yellow page category "Furniture Retailer", the number of keywords that can be recognized greatly expands. Now, a user can search for "Beds", "Bookshelves", and many other terms and be taken to a relevant list of businesses that sell these products. Taxonomy expands the types of queries that a user can get results for in a directory search environment.

This is just a partial list, of course. I'll have more posts like this in the future with other ways that taxonomy can be used.