0

AN OVERBROAD PATENT ON determining user opinions related to a product or service from online contents - This application from Metavana seeks to patent the idea of...Identifying keywords from social media contents, determining a category related to the identified keywords, and determining sentiments associated with the identified keywords! 10 minutes of your time can help narrow US patent applications before they become patents. Follow @askpatents on twitter to help.

QUESTION - Have you seen anything that was published before 9/30/2011 that discusses:

  • Determining sentiments expressed in social media contents

If so, please submit evidence of prior art as an answer to this question. We welcome multiple answers from the same individual.

EXTRA CREDIT - Category refers to products, people, hotels, events, services, places, or real world objects. Social media content refers to contents from websites (for e.g. amazon.com, bestbuy.com, yelp.com, tripAdvisor.com etc.), blogs, micro-blogging sites .

TITLE: Providing opinions about various categories based on sentiments expressed in social media contents

Summary: [Translated from Legalese into English] Identifying keywords in a sentence from social media contents, identifying polarity of the identified keywords using phase transition formula, determining categories related to the identified keywords, and determining sentiments corresponding to the categories related to the identified keywords.

  • Publication Number: US 20130091117 A1
  • Application Number: US 13/632,011
  • Assignee: Metavana
  • Prior Art Date: Seeking prior Art predating 9/30/2011
  • Link to Google Prior Art Search - "Find Prior Art"

Claim 1 requires each and every step below:

A computer-implemented method for sentiment analysis from social media content, comprising:

  1. Crawling, by a processor, a plurality of websites to obtain metadata from social media content;

  2. Extracting, by a processor, the metadata from the social media content by identifying a polarity of the sentiment-bearing keywords in a sentence from social media content using a phase transition formula;

  3. Determining at least one category corresponding to the at least one sentiment-bearing keyword of the sentence; and

  4. Determining at least one sentiment corresponding to the at least one category based on the at least one sentiment-bearing keyword.

In English this means:

A method for analyzing sentiments expressed in social media contents, comprising:

  1. Visiting a plurality of websites to obtain metadata from social media contents;

  2. Identifying polarity of keywords that express sentiments in a sentence from social media contents;

  3. Determining categories related to the keywords; and

  4. Determining sentiments associated with the determined categories based on the keywords.

Good prior art would be evidence of a system that did each and every one of these steps prior to 9/30/2011

You're probably aware of ten pieces of art that meet this criteria already... separately, the applicant is claiming Calculating a number of online documents that include a keyword, and number of instances of the keyword in the online documents, to determine relevancy of the keyword

"Determining public opinion based on social media contents" from the Applicant

What is good prior art? Please see our FAQ.

Want to help? Please vote or comment on submissions below. We welcome you to post your own request for prior art on other questionable US Patent Applications.

Improve this question

1 Answer 1

Reset to default
2

SAS Inc offer a product name 'SAS Sentiment Analysis' which was an acquisition of a company called 'Teragram Corporation'.

This product won the Business Wire product of the year award for its excellence in innovation. Below is an extract of from the press release, which describes the SAS Sentiment Analysis feature set / benefits which sound very similar to what the Metavana patent details:

SAS Sentiment Analysis’ high-performance crawler locates and extracts sentiment from digital content sources, including mainstream websites, social media outlets, internal servers and incoming news feeds.

In terms of how the crawled content is processed to locate sentiment it states:

SAS’ unique hybrid approach combines powerful statistical techniques with linguistics rules to improve accuracy to the detailed feature level. It summarizes the sentiment expressed in all available text collections – identifying trends and creating graphical reports that describe the expressed feelings of consumers, partners, employees and competitors in real time.

The above description makes the products sentiment analysis capabilities seem more advanced than the categories and keyword approach detailed in the patent application.

I researched further to see if I could find more information on the SAS product related to categories and keywords as a method to locate sentiment and found:

1) a post on the SAS blog called 'Predicting American Idol results based on Twitter sentiment' which details that keywords can be used. The post it states:

Generating the scores was easy enough with SAS Sentiment Analysis. I used a SAS macro to copy two weeks of Tweets about the final six contestants. From there, I took an existing Sentiment project (about an actual business), and copied over the list of generic positive and negative keywords. I ran the model and then added in a few extra rules to account for the nuances of Twitter-speak (turns out most Twitter-speak consists of shortenings of "love" and "sucks").

2) an article titled 'SAS Text Analytics and Teragram', which refers to the 'Content Categorization function' as described below:

Another important point is that SAS Text Analytics includes four components. There is the SAS Enterprise Content Categorization function. The system parses content and identifies entities. Metadata are created along with category rules. The second function is SAS Sentiment Analysis. A number of companies are competing in this sector. The SAS approach sucks in emails, tweets, and other documents. The system identifies various subjective shades in the source content.

I don't know the SAS product intimately but it very much appears to perform everything described in the patent application. The sources I have used are not perfect and they describe the SAS product from the user point of view as opposed to an informed objective one.

It actually looks like the SAS is far more advanced than what the patent describes which appears quite rudimentary.

One thing to note: I couldn't find any SAS or Teragram patents to use as sources.

Business Wire Press Release July 26, 2010

SAS Blog describing keywords as a component of locating sentiment May 12, 2011

Article describing categories as a component of locating sentiment May 28, 2010

Improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .