Techsplained: How Alexa, Siri, Google listen you to provide search results

By: |
Published: March 30, 2020 3:40:43 AM

As IBM Watson has been able to achieve some scale, the company believes that by putting the NLP technology in commercial use it will be able to expand its capabilities much further.

The specific branch fo AI that is used is called natural language processing.

Ever wondered how Alexa can understand you or Google can decipher what you are saying and provide search results. Or how Siri can tell you a joke once you ask for it/. Although artificial intelligence has a lot to do with how accurately tech companies have been able to decipher what you are trying to communicate. The specific branch fo AI that is used is called natural language processing. Although still in its initial phases of development, the technology is rapidly evolving. IBM recently announced that it would be furthering the services of its natural language processing programme to business users.

What is natural language processing and how does it work?

Natural language processing is how we interact with a machine. The primary job of a machine is to understand and make sense of what it is being told. The device uses algorithms based on a set of commands or conditions to do this. So, it is taught the basics of a language—the grammar, the syntax—and then left to determine how to perceive it. The most basic application of this has been the spell checks and grammar checks that a Microsoft Word runs. A bit more advanced version would be personal assistants.

So, what are the techniques used for natural language processing and why is it difficult?

The reason computers have not been able to learn natural language processing or NLP is because language is complicated, and there is no one way to decipher it. For instance, even the most sophisticated software of language can err on grammar as they can perceive a sentence differently than a human would. Two techniques are followed in language processing to improve results. One is called syntactic analysis. This is an entry-level language processing technique, where the computer determines if the natural language follows the grammatical rules. The algorithms apply such laws to a sentence and try to derive meaning from it. The other method is a bit more complicated and is called semantics. This has more to do with the structuring of sentences and making sense of them.

What is IBM doing?

IBM has used its artificial intelligence system called Watson to create Project Debater. Debater can scan through a multitude of documents to understand what are the essential points and then argue a case. It has also beaten a human in a debating competition. And, like any machine learning system as it learns more, its capabilities increase. The system can determine if two sets of an argument are similar even if they use a different language.

So, why is IBM using it in business?

As IBM Watson has been able to achieve some scale, the company believes that by putting the NLP technology in commercial use it will be able to expand its capabilities much further. As more organisations use it, IBM Watson will be processing much more data and learning more. What this means is over years the probability of your machine misinterpreting what you said will decrease. But, still, this does not mean that Watson will be able to conjure up a new idea and a new line of argument all by itself.

Techsplained @FE features weekly on Mondays. If you wish to send in queries that you want explained, mail us at ishaan.gera@expressindia.com

Get live Stock Prices from BSE, NSE, US Market and latest NAV, portfolio of Mutual Funds, calculate your tax by Income Tax Calculator, know market’s Top Gainers, Top Losers & Best Equity Funds. Like us on Facebook and follow us on Twitter.

Financial Express is now on Telegram. Click here to join our channel and stay updated with the latest Biz news and updates.

Next Stories
1Maharashtra deploys tech to fight Covid-19
2First response: Fighting the coronavirus with tech and team work
3How Marico Foundation helping innovators to create ingenious PPE