Everything You Need to Know About Text Mining

Written in collaboration with Mary D.

Text mining is widely recognized by companies as one of the major tools that are provided by A.I. technology to extract valuable ‘structured” data from text and help businesses filter through and condense valuable project-oriented information. If this sounds a little too ‘sci-fi” for you, that is exactly the point. Its capacity to recognize and respond to human speech and mimic the neural pathway activities of the human brain to develop independent cognitive abilities and behavioural responses is only the beginning, especially as more technological advancements are introduced into the field and artificial intelligence software are gaining higher momentum and demand among multiple industries with connections to the IT/Computer/Mobile industry—from the transport, banking, social services, government, and medical sectors among many more. While the first attempts to develop intelligent thinking machines can be traced historically to Raymond Lull in the 14th Century—and “automatons” are present in the ancient mythology of Greco-Roman, Egyptian and Babylonian/Mesopotamian literature—text mining goes back to the WWII era, when governments started adopting “content analysis” and assigning numerical codes to public concepts and ideas that were found in the media, including newspapers, magazines, letters, documents…etc. with the objective of analyzing and monitoring trends in mass-behaviour by tracking the levels of popularity and development of those concepts/ideas. This practice has also developed into another branch known as open source intelligence, used by governments and the intelligence communities to sift through all the pertinent information that is available on the World Wide Web for reasons of national security and especially in response to the current major crises facing the modern world. Unfortunately for the general public, it is also something that has led to the infringement of personal security and civil rights as demonstrated by its current use as a global spy network set up by agencies like the NSA and the CIA.

2013_11_060021d - (t3)

A.I.’s ability to extract concepts out of written language text through text analytics by analyzing and processing all the data present on the world wide web—including social media sites like Facebook, Twitter, LinkedIn, or YouTube, as well as independent professional blogs and websites—has enriched a company’s potential to accelerate decision and policy-making, while restructuring also its internal organization, reducing budget costs by saving time and being able to respond faster or anticipate any internal or external crises (such as changes in consumer trends or demand, or meeting important deadlines for a business venture). Even more importantly, it enhances customer engagement by delivering a unique experience that is able to identify and predict their tastes and cater to them individually and accordingly.

Its value for the general public is particularly evident in its applications in the medical industry, where A.I. will be able to monitor a person’s health and predict a heart attack days or even years before it occurs and respond appropriately by administering therapeutic treatment on the spot. The main difference compared to modern day search engines is that text mining can help a business or an individual find a solution for a need they did not even know existed! It can therefore help you find innovative sources that provide the answer to your problem before you even know a problem exists, and that, once again, is science fiction in the making! Integrated A.I. assistants deployed through a user interface on all your media devices including portable smartphones and tablets, and Robot receptionists/waiters are a perfect example of what we can expect in years to come. Sentiment Analysis, also referred to as “Opinion Mining”, for example, is another process whereby A.I. is able to analyze, extract and understand the emotional response of the subject in a context. In other words A.I. is truly the realization of our collective sci-fi archetypal imagery, as any fan of Star Trek would surely know!

The Promising Future of Artificial Intelligence

The following is a guest post

Artificial Intelligence (AI) has been a hot topic in the tech and innovation world as of late.  It has fueled the stuff of great sci-fi movies for generations, but only now is gaining traction in real, marketable products and services.

Yet, video games that feature AI aren’t particularly appealing; at least, not yet anyway. AI is functional, yet still lacks the flexibility and common sense of a real human. This is evident in plenty of online multiplayer games, where most computer-generated characters lack human reactions in complex situations.

Initially, Google’s Atari-playing algorithm is said to be the future of AI, and that we can learn a lot from its mistakes. However, just like many other artificial intelligence of the past, the platform and software wasn’t able to understand what’s going on in the game in the way a human does.

However, recent news revealed that Google’s AI was able to win  the fifth and final game against Go genius, Lee Sedol. The computer system was able to defeat the Korean grandmaster with one loss, marking a significant moment for artificial intelligence. Although machines have beaten the best humans at checkers, chess, and even Jeopardy!, it is the first time a machine was able to top the very best at Go.

Reliving icons

Exceeding video games, AI has been able to connect us with musicians of the past that have sadly left us prematurely.

Tupac Hologram at Coachella 2012

We have seen how AI has been used to bring musicians back from the dead as reported by Popsci, from Tupac in a recent Coachella performance to other memorable artists such as Jimi Hendrix and Kurt Cobain. Software created in 2010 works like ‘a Pandora for live music’ as it analyzes a musician’s voice and sound based on their old, archaic recordings. It will then reconstruct a song using those musicians’ voices as if they have recorded in a modern studio. Thus, it becomes easier for modern musicians to collaborate with their favorite music icons.

Reliving the lives of past musicians is not at all a new concept. In fact, many mobile and online games have been inspired by these icons, such as the Jimi Hendrix slot game that takes you back in time with its groovy 60s design and concepts akin to the popular Guitar Hero game which has spawned offshoots such as the Metallica version for console platforms. But, it’s not only musical legends that have games tailored for them, as recent celebrities also make their own titles from Kim Kardashian to NBA players. Though, artificial intelligence turns these games into a whole different experiences – transforming them into a more immersive, realistic, and engaging proposition entirely.

Andrew Moore, dean of CMU’s School of Computer Science, said AI requires large volumes of data and statistical computation to function, making it difficult to easily produce. However, it can unlock plenty of new avenues for technology in the future.

“Teaching AIs to process scenarios with hidden information will unlock whole new vistas of applicability for the technology.” — Andrew Moore

Doctor AI

A sector that is said to benefit from AI is the medical industry, as experts suggest they are now looking at AI 3D hologram avatars to help care for the elderly.

“Although this project is at an early stage, with a number of technical, moral and ethical issues to be addressed, the development of Rita (artificial intelligence) in the form of a humanised avatar could revolutionise how an individual’s personal, social emotional and intellectual needs are met in the future,” said co-director of the University of Kent’s Centre for Child Protection (Dr Jane Reeves in an interview with International Business Times).

Photo Credits: evsmitty via Compfight cc