Index update: Topics

It’s time for another Index update!

Over here at Index HQ we’re always hard at work, plugging away to add useful features for all your tech-scouring needs.

Index runs on news. Every day we push over 1,000 articles through the system’s ever-hungry maw and match them to companies and locations in our database. News is the context layer that informs and enriches the data we gather.

214245 no5 reading 897a61 original 1466171381

But simple processing isn't enough. The average article has so much more to offer between the headline and author bio than raw data alone. As we continue to find ways to teach Index to read between the lines, we build features that utilize the stories we find there.

Our first step in that direction is topics. Topic detection on Index works through natural language processing and supervised machine learning. By processing news articles we determine what the main topic of the text is, and label it accordingly.

214251 screen%20shot%202016 06 17%20at%2016.14.55 bf78c7 original 1466173188

We’re taking a classic iterative approach here, and releasing small updates as we go. For now, topics are purely informative – we’ll be labeling articles in your news streams with their respective topics, but we’re not yet collecting them in one place.

Before we move forward, we want to see what the community thinks of the topics. Which do you find useful, which seem redundant? Which are spot-on and which are completely inaccurate?

We’ll be keeping you updated on further developments. If you have any ideas on where you’d like to see topics go from here, let us know! Get in touch on or shoot us a message at

Download PDF
Download PDF
About Index

Index provides insights in private tech companies by turning unstructured content into structured data and intelligence. A tool for the tech community, tech companies, journalists, investors and analysts.