AdsCollect - Dhaka

AdsCollect is a start-up that crawls digital versions of newspapers to detect ads. These digital versions of newspapers are published on AJAX-based interfaces, which make it difficult for conventional search engines to crawl and index them. To combat this issue, AdsCollect uses machine learning to detect ads, followed by Optical Character Recognition (OCR) to provide the contextual information of the ad. Natural Language Processing (NLP) is then used to index the text obtained by OCR. Through this, AdsCollect creates an index of the digital versions of newspapers which are otherwise invisible on conventional search engines.

AdsCollect offers a free service that relies on the "adwords for newspaper" business model. Its current goal is to bring an intuitive experience to print media  

Show more

advertisers, similar to running a campaign on Adwords or Facebook. With this, searching for tender notices or job circulars published in the newspaper becomes much easier.