Many clients and partners have asked me about when the Personality Insights service will become generally available. The wait is over!
Today, we are pleased to announce the General Availability (GA) release of the IBM Watson Personality Insights service (formerly known as the IBM Watson User Modeling service in Beta). Like the other Watson services, Personality Insights is available on Bluemix, IBM’s open cloud development platform.
Here’s a brief introduction to what the Watson Personality Insights service is all about:
The Personality Insights service uses linguistic analytics to infer the personality characteristics, intrinsic needs and values of individuals from communications that a user opts to make available via mediums such as email, text messages, social media, forum/blog posts, and more. These insights help businesses better understand their clients and improve customer satisfaction by anticipating customer needs and recommending future actions. This allows businesses to improve new client acquisition, retention, and engagement, and strengthen their relationships with existing customers.
The following is a brief description of the three kinds of personality insights that are provided by this service:
- Personality characteristics: The service can build a portrait of an individual’s personality characteristics and how they engage with the world across five primary dimensions: Openness, Conscientiousness, Extroversion, Agreeableness, and Neuroticism (also known as Emotional Range).
- Needs: The service can infer certain aspects of a product that will resonate with an individual across twelve needs: Excitement, Harmony, Curiosity, Ideal, Closeness, Self-expression, Liberty, Love, Practicality, Stability, Challenge, and Structure.
- Values: The service can identify values that describe motivating factors which influence a person’s decision-making across five dimensions: Self-transcendence / Helping others, Conservation / Tradition, Hedonism / Taking pleasure in life, Self-enhancement / Achieving success, and Open to change / Excitement.
IBM Watson Personality Insights Updates since Beta
After working with clients who used the beta version of this service, we received valuable feedback that led to new enhancements, features and functional improvements that are delivered in the GA release of the Personality Insights APIs, such as:
- We added a textual summary of personality characteristics. Clients can use this description as-is or modify it when integrating the Personality Insights service into their business applications. The following is a sample text summary from the sample application:
- We enhanced the service to accept two new input formats, adding plain text and HTML inputs to JSON format that was already available in the profile API. These new types can be selected via the “Content-Type: text/plain” and “text/html” headers.
- We enhanced the profile API to return a new output format, adding CSV output to the profile API. This output format can be selected via the “Accept: text/CSV” header. It also adds an optional “headers” parameter to the profile API call.
- We provide many improvements to the ‘Needs’ model. To improve the sampling error of Needs characteristics, we updated the needs modeling by retraining with more ground truth data, and using additional dictionaries. We recommend that any results generated by the beta APIs should be recomputed.
- We added a sampling error score. Language use varies naturally from document to document, and time to time. Given this variation, a small sample of text may not be representative of an individual’s overall language patterns. However, in some cases, users may only have a small amount of text from an individual and may still want to derive insights about that individual. In such cases, we understand that it’s important to know the sampling error associated with the percentile scores of characteristics that our service outputs. Therefore, we added a new output element to the profile API named Sampling Error Score. Because characteristics are reported in terms of population percentiles (which means that to say that a person scores in the 37th percentile means that he/she is higher in the score for that characteristic than 37% of the population), our API reports the sampling error in terms of percentiles. Sampling error can be reduced by analyzing more text from an individual: the more text that you provide, the lower the sampling error, and vice versa.
These improvements are backwards compatible and are not expected to affect applications that are currently using the Personality Insights APIs. These improvements provide more flexibility and deliver improved analytic results.
Pricing and Freemium Tier
Trial Bluemix accounts remain free. If you don’t already have an account, you can go to www.bluemix.net, to register, and get free instant access to a 30-day trial without a credit card. Use of the Personality Insights service is free during this trial period. After the 30-day trial period, you can access the Personality Insights service free of charge for 100 API calls per month. If you need to make additional calls, you will be charged at a rate of 60 cents/API call. Additional Bluemix discounts may apply. Please check bluemix pricing here.
We are looking forward to continuing to work with the many clients, business partners, and creative developers that built innovative applications using the beta version of the Personality Insights service. If you have done so, please migrate your applications to use the GA service in the next 30 days (by March 24, 2015), after which the beta User Modeling service will no longer be available. We’re eager to see what the next round of cognitive applications based on Personality Insights look like!
Personality Insights and our Partners
See how our partners are already using the Personality Insights service. Read more about eyeQ and how they’re improving the in-store shopping experience in the modern market.