point

HINDI: PENETRATING THE DIGITAL SPACE

By Peeyush Bajpai
Co-Founder-Raftaar.in
Friday, January 10, 2014
By Peeyush Bajpai
Founded in 2005, Raftaar is a search and exploration Hindi online portal that has over 2.5 million unique visitors per month. The portal allows users to type in Hinglish, and Hindiusing a customized keyboard available on the platform.

Access to information is an integral part of our being. The RTI (Right to Information) act and the zeal with which it has been adopted by ordinary citizens in India is now a legendary folklore. The power of being able to access information and use it to better lives has only further strengthened the mediums that can enable this. Internet by far is one such medium that not only enables access to information but also helps to communicate and engage with other citizens in a cost effective manner. The world over this revolution has created a boom in flow of content. In India too content has increased, but the growth was muted due to lack of Indian language solutions. To be able to effectively reach this content in the fastest available time, led to the development of search technology.

Today, most of us accessing the internet are all attuned to having instant information courtesy search engines. But within India, with the dialectical variations, the success rate of search engines has not been as enormous. The obvious facts are the lack of internet penetration but more importantly, lack of standardization of Indian language solutions. We are all sentient of the myths surrounding Hindi as a language. Hindi is spoken by almost 500 million people across the globe, making it the third most spoken language in the World. A high context language, it has evolved from the Brahmi script and has acquired its current form over many centuries, and numerous dialectal variations still exist. Like Sanskrit, Hindi is written in the Devnagari script, which is common to couple of other Indian sub continent languages as well.

Over a period of time with the evolution of internet, Hindi has penetrated the digital space. With increasing number of people preferring to access content in local languages, there is a growing need to make content available in Hindi. Hence, with this initiative the pioneers in search, aggregation, and exploration of Hindi digital content, Raftaar.in has bridged the gap between Hindi content creators and users. The aim of search has always been to provide the most relevant result to a user. This increases the usability of internet and gets more users to come more often. As more traffic grows, more content would be generated to meet the various needs of users. However, this is easier said than done. Content has grown manifold with various versions of same content floating over the internet. Further, every user has a different intent while each may use the same term to search. The context which led the user to search is not available to the search engine. All this has pushed search to evolve from not just returning pages from the net containing the terms to providing ways to explore information. Google, the market leader in search has been processing terabytes of data to develop an intent engine and a knowledge graph. It is trying to determine how search query resulting from a conversational communication be best matched with documents that are more for a written form of communication. For the English language, much research has been done and outputs implemented. Thus at Raftaar.in, the current focus is on creating implementable solutions from natural language processing research which would provide better insights into the context of any search query rather than just the meaning of words being entered.
Today�s standard user is inquisitive about the intuitive thinking that goes behind making this querying process simpler, faster, and more importantly, customized. The users have a history and are experienced of the powers of the medium. The digital medium is responding by making digital conversations more intuitive in nature. The non-English audience while being more aware of the technical advances however is more conservative when it comes to content and hesitates to explore the vernacular world. Understanding search behavior is thus very important for a non-English search engine. But even when the audience is English savvy as in Urban India, the fact that users may learn English but do not think in English, is crucial on designing the service. An unequivocal fact for countries who do not have English as their first language. Raftaar.in has effectively created and developed applications to enhance user experience keeping in mind the bilingual ability of the audience.

It can be safely assumed in India that the next wave of internet usage can only come from vernacular space, language focused search is therefore now uninhibitedly hot property. It is also a critical component for the entire ecosystem as it bridges the niche gap between a localized user and his associable language specific content. Technology firms need to now respond in a manner that technology becomes an enabler from a digital barrier that it became. Raftaar's innovative solutions have been developed with this in mind. The current solutions separates the Hindi content from the entire web, indexing, categorizing and classifying it in an automated manner on a real time basis and making the otherwise difficult to access information easily available. Its unique classification algorithm combined with usage of analytics enables the content into specific easy to access verticals.

Share on Twitter
Share on LinkedIn
Share on facebook