Senior Developer at Website Categorization API startup
Joined November 2020
- Tweets 19
- Following 22
- Followers 0
- Likes 51
another post, on #predictiveanalytics kategorisierungen.substack.c…
1
a couple of useful links on the topic website #categorizations:
linktr.ee/categorizationlink…
1
Websitecategorization retweeted
$300K+ per year with an 85% margin. Without paid ads.
That's my Shopify app WideBundle.
Here's the exact method I'd use if I had to start again: 🧵👇
34
79
5
589
Websitecategorization retweeted
#8 Don't build something new
Find an industry that already has the product-market fit.
The reason for this:
Your chances of success are astronomically higher when you enter a market that already has product-market fit.
5
15
151
New Saas product for #ai #product #categorization based on #nlp analysis:
productcategorization.com/de…
1
the service is using ML model trained on large number of annotated product texts. google product taxonomy was taken as the base taxonomy, which was then customised.
1
the #platform can be used both as dashboard or as #API #json endpoint. here is the url for IAB classification that you can try out: productcategorization.com/de…
1
the new platform is a result of almost a year of development, mostly on acquisition of high quality #datasets for product texts, as well as developing custom product #taxonomy.
@threadreaderapp unroll
2
also added to the service is the #NMT service, which allows us to offer #textclassification for 100+ languages. we will soon also add new subscription plans that cover all needs of users: productcategorization.com/pr…
an important part of website categorization is extraction of texts from website. i.e. separating boilerplate content, like menus, footers from main content. there are many python libraries that work well for this task, like goose3 and newspaper
we have added the feature of #classifying general texts (in addition to URLs) according to either custom Google Product Taxonomy or IAB:
websitecategorizationapi.com…
Websitecategorization retweeted
Social media sentiment of #bitcoin was cooling already hours before the fall of $BTC, see chart. Sentiment determined by applying #machinelearning on social posts about $BTC.
There were also extremely weak support levels between 54k and 50k, see chart. bittsanalytics.com
1
2
3
