November 19, 2013
Sad to say, we have heard rumblings about severe disappointment with Attensity-type and Lexalytics-type sentiment applications. If you want to kick some tires in this interesting search niche, look instead to the open source application TextBlob. OpenShift points out this resource in, “Day 9: TextBlog—Finding Sentiments in Text.” The article is one in an ambitious series by writer Shekhar Gulati, who challenged himself to master one technology a day for a month. Very admirable, sir!
Gulati begins with his experience with sentiment analysis:
“My interest in sentiment analysis is few years old when I wanted to write an application which will process a stream of tweets about a movie, and then output the overall sentiment about the movie. Having this information would help me decide if I wanted to watch a particular movie or not.
“I googled around, and found that Naive Bayes classifier can be used to solve this problem. The only programming language that I knew at the time was Java, so I wrote a custom implementation and used the application for some time. I was lazy to commit the code, so when my machine crashed, I lost the code and application. Now I commit all my code to github, and I have close to 200 public repositories
“In this blog, I will talk about a Python package called TextBlob which can help developers solve this problem. We will first cover some basics, and then we will develop a simple Flask application which will use the TextBlob API.”
The post does indeed cover the basics, including the installation of Python and virtualenv before we can get going with TextBlob. It then takes us through writing an example application and deploying to the cloud. As he notes above, Gulati has his code safe and sound at Github; the code for this example are posted here, and the js and css files can be found here.
Cynthia Murrell, November 19, 2013
November 2, 2013
Simon Creasey from Computer Weekly recently reported on the outcome of the latest Twitter firestorm in the article “Failure to Invest in Sentiment Analytics Could Lead to Brand Damage.”
According to the article, a disgruntled British Airways passenger decided use a paid-for promoted tweet to blast his complaints to thousands of Twitter followers. As you can imagine, the tweet went viral and was shared and re-shared until it received global coverage. While PR disasters are often unavoidable, businesses are developing social media sentiment analysis software to contain them.
The article concludes:
““Monitoring what people are saying about your products and industry can help you design your products and propositions for the future and in that sense Twitter acts as a great market research tool as well as a lead-generation tool,” says Sinclair.
“Similarly, if you monitor what people are saying about your brand it can also help you with customer service and PR. There are many examples of companies who have found themselves under social media attack. Failure to invest in these kinds of tools could easily result in significant damage to a company’s reputation and brand.”
These days, social media is ever expanding and it is impossible to keep track of everything being said about your company’s brand, products, and employees. In order to avoid PR disasters like the one that happened to British Airways, companies should invest in the latest sentiment analysis technologies.
Jasmine Ashton, November 02, 2013
October 21, 2013
Here is something new from Gigaom: “Stanford Researchers To Open Source Model They Say Has Nailed Sentiment Analysis.” Richard Socher and a team from Stanford have created a computer program that can classify the sentiment of sentence with 85% accurately. They tested the model on movie reviews with a positive or negative tone. Even more amazing is that Socher and his team are making the project available to everyone. Why not capitalize on it instead? After all, companies have been trying for years to analyze social media and would pay the big bucks for said technology.
What makes Sucher’s project different from other sentiment software is that is reads whole sentences rather than just words.
“The team then built a new model it calls a Recursive Neural Tensor Network (it’s an evolution of existing models called Recursive Neural Networks), which is what actually processes all the words and phrases to create numeric representations for them and calculate how they interact with one another. When you’re dealing with text like movie reviews that contain linguistic intricacies, Socher explained, you need a model that can really understand how words play off each other to alter the meaning of sentences. The order in which they come, and what connects them, matters a lot.”
Socher hopes to reach a 95% accuracy, but the technology will never be 100% accurate because of jargon, idioms, odd word combinations, and slang. The project is making landmark strides in machine learning, logical reasoning, and grammatical analysis.
It means better news for online translators and speech technology, but commercial sentiment analytics vendors may see a decline in their profits.
Whitney Grace, October 21, 2013
August 20, 2013
Calling all software developers, analysts and systems integrators. The leading semantic intelligence developer, Expert System is hosting a webinar entitled, “What’s Hiding In Your Data? Test Drive Our Semantic API.” The webinar is scheduled for August 28 at 12 pm ET/9 am PT and registration is now open.
We recommend that professionals who are interested in transforming content and data streams into actionable and strategic information should sign up. A unique offering of this webinar is the live product test drive so that those interested can see how their flagship Cogito Intelligence API works.
The webinar description summarizes Cogito Intelligence API:
Cogito Intelligence API is a unique API that uses the power of semantic processing—Text Mining, Categorization, Tagging—and deep domain vertical knowledge for Intelligence to help analysts access and exploit some of their most strategic sources of information. As the only semantics based system, Cogito Intelligence API provides complete understanding of meaning and context in the processing of data and resolves ambiguities in data more effectively than solutions based on keywords or statistics.
Another unique offering from the Cogito API revolves around corporate security. Their solution is already embedded with corporate security measures, which enables businesses to operate all applications with the same confidence that Cogito offers.
Megan Feil, August 20, 2013
August 13, 2013
When enterprise organizations understand the value of unstructured data, and especially the value of it when it is integrated with structured data, what kind of solutions do they utilize? According to a recent study by Altimeter Group reported in “Enterprise Social Data Isolated in Departmental Silos,” 42 percent of the 35 large organizations surveyed were using business intelligence tools. Other areas where data sets converged were market research at 35 percent, CRM at 27 percent, email marketing at 27 percent and sensor data at four percent.
The main argument presented by the article is that as long as people are working in departmental silos, information and data will be first and foremost stored in a way that parallels how people are organized.
We learned more about why some organizations face challenges when integrating data:
The report also revealed it’s not always easy to integrate this data, attributing the issue to the fact that so many organizational departments touch the data, ‘all with varying perspectives on the information,’ the article states, adding: ‘The report also notes the numerous nuances within social data make it problematic to apply general metrics across the board and, in many organizations, social data doesn’t carry the same credibility as its enterprise counterpart.’
We know that one company, Expert System, would have quite the rebuttal to this argument that unstructured data may not be worthy across the board for all departments. Their solution Cogito Intelligence API yields insights and actionable information after parsing both structured and structured data while using sentiment analysis and natural language processing technologies.
Megan Feil, August 13, 2013
August 1, 2013
We are seeing a lot of information published in regards to the ties between search and collaboration. As collaboration inherently relies on search, it is no wonder that these two are consistently discussed in tandem — “How Search Amplifies Enterprise Collaboration” from Business2Community points this out too.
This article discusses how social features and metadata make information more findable and thus more likely to be used in collaborative projects between users.
The author, Christian Buckley, explains his evolving perspective on sentiment analysis:
I questioned the ability of this technology to adequately interpret and intelligently map end user sentiment to content and metadata, or “data about data,” improving the overall search experience. Sentiment analysis is an incredibly difficult thing to automate, much less deliver within mainstream platforms. Thankfully, we have a method for providing a robust sentiment-based layer to our structured collaboration platforms: social collaboration. Even the search leaders recognize that they cannot completely replace human interaction (at least not yet) as the ultimate semantic classification mechanism.
Collaboration is one key reason companies are seeking out enterprise search vendors utilizing semantic technologies. Expert System is one such company whose solutions offer precise analytics using their core semantic search technologies. Their linguistic analysis capabilities enhance the extraction and application of data in the natural language interface. Collaboration is only the beginning, Expert System also has semantically enriched tools for social media monitoring, customer service and more.
Megan Feil, August 1, 2013
July 23, 2013
Looking for an alternative to Google and other big-name web search platforms? Teach Amazing recommends “Hakia—Semantic Search Portal.” Writer Mark Brumley explains that the semantic web search portal presents multiple types of content in the same results page. The various sections encompass the gamut and this type of aggregated search displayed on a single point of access seems to be the wave of the future.
Brumley shares his user experience:
I love how search results are displayed. Various sections are populated on a single page. Sections include web, news, blogs, Twitter, images and videos. The Twitter feed is a real plus for me and gives an indication of the current pulse of a particular topic. Give Hakia a try the next time you are doing some research. Make sure you try it in the classroom as well. Students need to know that Google is not the only search provider on the planet.
Brumley is not alone in recognizing the advantages of parsing data with context and meaning that semantic search provides. The enterprise also functions more efficiently when using tools that take a semantic approach to data. For example, Expert System offers solutions that empowers users to work at new heights of discovery and collaboration.
Megan Feil, July 23, 2013
May 29, 2013
Identifying user sentiment has become one of the most powerful analytic tools provided by text processing companies, and Bitext’s integrative software approach is making sentiment analysis available to companies seeking to capitalize on its benefits while avoiding burdensome implementation costs. A few years ago, Lexalytics merged with Infonics. Since that time, Lexalytics has been marketing aggressively to position the company as one of the leaders in sentiment analysis. Exalead also offered sentiment analysis functionality several years ago. I recall a demonstration which generated a report about a restaurant which provided information about how those writing reviews of a restaurant expressed their satisfaction.
Today vendors of enterprise search systems have added “sentiment analysis” as one of the features of their systems. The phrase “sentiment analysis” usually appears cheek-by-jowl with “customer relationship management,” “predictive analytics,” and “business intelligence.” My view is that the early text analysis vendors such as Trec participants in the early 2000’s recognized that key word indexing was not useful for certain types of information retrieval tasks. Go back and look at the suggestions for the benefit of sentiment functions within natural language processing, and you will see that the idea is a good one but it has taken a decade or more to become a buzzword. (See for example, Y. Wilks and M. Stevenson, “The Grammar of Sense: Using Part-of-Speech Tags as a First Step in Semantic Disambiguation, Journal of Natural Language Engineering,1998, Number 4, pages 135–144.)
One of the hurdles to sentiment analysis has been the need to add yet another complex function which has a significant computational cost to existing systems. In an uncertain economic environment, additional expenses are looked at with scrutiny. Not surprisingly, organizations which understand the value of sentiment analysis and want to be in step with the data implications of the shift to mobile devices want a solution which works well and is affordable.
Fortunately Bitext has stepped forward with a semantic analysis program that focuses on complementing and enriching systems, rather than replacing them. This is bad news for some of the traditional text analysis vendors and for enterprise search vendors whose programs often require a complete overhaul or replacement of existing enterprise applications.
I recently saw a demonstration of Bitext’s local sentiment system that highlights some of the integrative features of the application. The demonstration walked me through an online service which delivered an opinion and sentiment snap in, together with topic categorization. The “snap in” or cloud based approach eliminates much of the resource burden imposed by other companies’ approaches, and this information can be easily integrated with any local app or review site.
The Bitext system, however, goes beyond what I call basic sentiment. The company’s approach processes contents from user generated reviews as well as more traditional data such as information in a CRM solution or a database of agent notes, as they do with the Salesforce marketing cloud. One important step forward for Bitext’s system is its inclusion of trends analysis. Another is its “local sentiment” function, coupled with categorization. Local sentiment means that when I am in a city looking for a restaurant, I can display the locations and consumers’ assessments of nearby dining establishments. While a standard review consists of 10 or 20 lines of texts and an overall star scoring, Bitext can add to that precisely which topics are touched in the review and with associated sentiments. For a simple review like, “the food was excellent but the service was not that good”, Bitext will return two topics and two valuations: food, positive +3; service, negative -1).
A tap displays a detailed list of opinions, positive and negative. This list is automatically generated on the fly. The Bitext addition includes a “local sentiment score” for each restaurant identified on the map. The screenshot below shows how location-based data and publicly accessible reviews are presented.
Bitext’s system can be used to provide deep insight into consumer opinions and developing trends over a range of consumer activities. The system can aggregate ratings and complex opinions on shopping experiences, events, restaurants, or any other local issue. Bitext’s system can enrich reviews from such sources as Yelp, TripAdvisor, Epinions, and others in a multilingual environment
Bitext boasts social media savvy. The system can process content from Twitter, Google+ Local, FourSquare, Bing Maps, and Yahoo! Local, among others, and easily integrates with any of these applications.
The system can also rate products, customer service representatives, and other organizational concerns. Data processed by the Bitext system includes enterprise data sources, such as contact center transcripts or customer surveys, as well as web content.
In my view, the Bitext approach goes well beyond the three stars or two dollar signs approach of some systems. Bitext can evaluate topics or “aspects”. The system can generate opinions for each topic or facet in the content stream. Furthermore, Bitext’s use of natural language provides qualitative information and insight about each topic revealing a more accurate understanding of specific consumer needs that purely quantitative rating systems lacks. Unlike other systems I have reviewed, Bitext presents an easy to understand and easy to use way to get a sense of what users really have to say, and in multiple languages, not just English!
For those interested in analytics, the Bitext system can identify trending “places” and topics with a click.
Stephen E Arnold, May 29, 2013
Sponsored by Augmentext
March 27, 2013
For a simple explanation of content enrichment, there is Web CMS Content Enrichment with OpenCalais, Crafter Rivet and Alfresco, on Rivet Logic Blogs. Content enrichment, the art of mining data and adding value to it, has now been organized by such services as OpenCalais, a free resource of semantic data mining from Thomson Reuters. For use on your blog, website or application, OpenCalais’s mission is to make “the worlds content more accessible.” The article explains,
“A few examples of content enrichment include: entity extraction, topic detection, SEO (Search Engine Optimization,) and sentiment analysis. Entity extraction is the process of identifying unique entities like people and places and tagging the content with it. Topic detection looks at the content and determines to some probabilistic measure what the content is about. SEO enrichment will look at the content and suggest edits and keywords that will boost the content’s search engine performance. Sentiment analysis can determine the tone or polarity (negative or positive) of the content.”
The tutorial on using OpenCalais with Crafter Rivet’s operating platform offered in this article is short and straightforward. Without tools like OpenCalais, the huge advantages of content enrichment for author and content managers would take countless hours. The resources available can save time while improving the effectiveness of content.
Chelsea Kerwin, March 27, 2013
March 20, 2013
Antonio S. Valderrábanos, founder of Bitext, recently granted an exclusive interview to the Arnold Information Technology Search Wizards Speak series. Bitext provides multilingual semantic technologies, with probably the highest accuracy in the market, for companies that use text analytics and natural language interfaces. The full text of the interview is available at http://www.arnoldit.com/search-wizards-speak/bitext-2.html.
Bitext provides B2B multilingual semantic technologies with probably the highest accuracy in the market. Bitext works for companies in two main markets: Text Analytics (Concept and Entity Extraction, Sentiment Analysis) for Social CRM, Enterprise Feedback Management or Voice of the Customer; and Natural Language Interfaces for Search Engines and Virtual Assistants. Visit Bitext at http://www.bitext.com. Contact information is available at http://www.bitext.com/contact.html.
Bitext is seeing rapidly growth, including recent deals with Salesforce and the Spanish government. The company has added significant and important technology to its multilingual content processing system.
In addition to support for more languages, the company is getting significant attention for its flexible sentiment analysis system. Valderrábanos gave this example: “flies” may be a noun, but also a verb. We say “time flies like an arrow” versus “fruit flies like bananas.” Bitext believes computers should be able to parse both sentences and get the right meaning. With that goal in mind, they started the development of an NLP (natural language processing) platform flexible enough to perform multilingual analysis just by exchanging grammars, not modifying the core engine.
He told ArnoldIT’s Search Wizards Speak:
Our system and method give us a competitive advantage with regards to quick development and deployment,” Valderrábanos said. “Currently, our NLP platform can handle 10 languages. Unlike most linguistic platforms, the Bitext API ‘snaps in’ to existing software.
Bitext’s main area of research is focused on deep language analysis, which captures the semantics of text. “Our work involves dealing with word meanings and truly understanding what they mean, interpreting wishes, intentions, moods or desires,” Valderrábanos explained. “We just need to know what type of content, according to our client, is useful for her business purposes, and then we program the relevant linguistic structures.” He added:
Many vendors advocate a ‘rip and replace’. Bitext does not. Its architecture allows our system to integrate with almost any enterprise application.”
Bitext already delivers accuracy, reliability and flexibility. In the future, the company will be focusing on bringing those capabilities to mobile applications. “IPads, tablet devices in general, and mobile phones are becoming the main computing devices in a world where almost everybody will be always online. This opens a new whole arena for mobile applications which will have to cater for any single need mobile users may have,” Valderrábanos said.
Donald C. Anderson, March 20, 2013