Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: NLP-based tool for technology research

researchly.app

Post reply on HN
Print view View original

11–16 of 16 posts

Re: Show HN: NLP-based tool for technology research

#11

Post by BLP4YC »

omeysalvi wrote: post #7

Who is the target market? Creators and inventors?

Please, tell me. Kidding, but also not. Honestly, I am still unsure. I started working on it because I came across those features in different papers (if I remember correctly - by accident) and thought they were cool. So I started working on it.

But "creators and inventors" goes in the direction I was thinking. Additionally, innovation consultants, tech. advisors, "policy makers" or people who have to decide on technology but lack technical expertise in certain fields, engineers...

Re: Show HN: NLP-based tool for technology research

#13

Post by Endlessly »

BLP4YC wrote: post #10
Endlessly wrote: post #6

Earlier quoted context omitted.

Curious, appears you used the same system for technical analysis of crypto ISOs valuations; if true, what if anything did you learn that to you was the most surprising as it relates to emergent tech know valuations?

Yes, absolutely right. The core tech. between crypto-valuations and tech-valuations stayed the same. It was basically: Python + spacy (plus a lot of different supporting libraries). Some surprising aspects: For tech-analysis, you can accomplish a lot by using a rule-based parsing because the source texts (e. g. patents) have the same sentence structure. In fact, several researches have shown that patent text follow a…

For keyword extractions, since you’re already using spaCy, you might take a look at textacy, specifically:

textacy.ke.utils.most_discriminating_terms

Documentation is here:

https://chartbeat-labs.github.io/textacy/build/html/api_refe...

Code is based on this research:

King, Gary, Patrick Lam, and Margaret Roberts. “Computer-Assisted Keyword and Document Set Discovery from Unstructured Text.” (2014). http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.458...

——

Thanks for the depth of your replies, love mapping information, finding patterns, “seeing the future”... though more often than not, even if you knew for sure some aspect of the future in a predictable way, still very hard to make use of it, mainly enjoy the topic as more as a info geek more than anything else.

Again, thanks for sharing!

Re: Show HN: NLP-based tool for technology research

#14

Post by Endlessly »

BLP4YC wrote: post #9
xtiansimon wrote: post #8

I've toyed with scripts from my NLP research, but nothing to this extent. Kudos. The webpage looks clean. The ratios of graphic elements to their scale is pleasant. The two-tone color palette is pleasant. The line length in the main screen (~140 c/l) is a bit much for my taste. As far as the Try now, I'm not sure how these relate to each other. Maybe a video demo walk-through taking one technology, such as the digita…

Thanks! I have been experimenting with NLP for two/three years now and this finally something somewhat useful. So your comment makes me really happay! Thanks for your feedback regarding the main page. I also like the design, but not because I made, but because the developers behind the templates did such a great job: https://startbootstrap.com/themes/sb-admin-2/ + https://blackrockdigital.github.io/startbootstrap-lan…

Agree with the feedback on being able to take a raw inputs such as even a manually selected set of documents such as wiki-pages, patents, research papers, etc — and see both the high-level workflow and even step-by-step the transformations and related code.

Clearly, that’s asking a lot both in terms, work and IP disclosure, but I am guessing what the average user would want, that is they have a specific need based on existing documents that the want ingested for analysis. Maybe I am wrong, but agree when doing a quick look, aside from the raw data & code, what I thought to look for to try and understand what was really there.

Re: Show HN: NLP-based tool for technology research

#15

Post by Endlessly »

BLP4YC wrote: post #11
omeysalvi wrote: post #7

Who is the target market? Creators and inventors?

Please, tell me. Kidding, but also not. Honestly, I am still unsure. I started working on it because I came across those features in different papers (if I remember correctly - by accident) and thought they were cool. So I started working on it. But "creators and inventors" goes in the direction I was thinking. Additionally, innovation consultants, tech. advisors, "policy makers" or people who have to decide on techn…

If I had to guess legal due diligence (acquisition, filings, processing-filings), competitive intelligence, large corporations with established R&D or IP efforts, financial intelligence, etc — would all be markets worth looking into.

Creators & inventors tend to want a lot of freebies in my experience.

Post reply on HN