New Approach to Automated Company Classification

blog.getcontext.io

11 points·vnture··6 comments

Hm.

It seems to work somewhat for SV startup companies, but fails pretty badly on others.

Eg:

IBM

https://www.getcontext.io/classifier/#/about/ibm: Evenly distributed between Augmented Reality/Cloud Computing/Cloud Infrastructure/Entertainment/Fintech/HR

Google

https://www.getcontext.io/classifier/#/about/google - apparently about virtual currency and crypto currency as much as a search engine.

Ford: No result

GE: Serves results for GEICO

Hi NL, thx for giving our hack a spin. We are working on it. Next version will have auto generated feature trees which will result in much better results.

Nice tool.

But as nl stated, not that accurate for some companies.

Would be great if you could describe in more detail how you create the classification.

thx, recognized the variance, two things: results are based on a tech domain specific vector (based on 1.2m tech articles / 20y), so non-techs fall off. We source content on requested entities live. Descriptions of younger entities are more concise/less global -> more valid classification. But then again, our assumption was that there is more demand for classification of little known entities than for F100s. But we are here to validate/falsify that assumption.

*vector space

Hey everyone! We have set up a demo, feel free to give it a try, all feedback is welcome! Thanks

Reply on news.ycombinator.com