Thanks for the feedback! I agree that the recognition of location entities is not accurate in some cases. For example if a text like "Blockai | San Francisco

Thanks for the feedback! I agree that the recognition of location entities is not accurate in some cases. For example if a text like "Blockai | San Francisco, CA | CV/ML and Front-end Engineers - https://blockai.com"(from https://blockai.com"(from https://news.ycombinator.com/item?id=12631335 https://news.ycombinator.com/item?id=12631335) is passed to Google Natural Language API: then it returns the following "locations": San Francisco, CA, ML, CV. https://github.com/sAbakumoff/hn-sort-out/blob/master/LangAnalytics.PNG https://github.com/sAbakumoff/hn-sort-out/blob/master/LangAn... However, ML stands for "Machine Learning" and CV is "Computer Vision". Perhaps the algorithm concludes that CV/ML is the location because it's close to other locations(San Francisco, CA). I will try to work it out for the next "Who's hiring" event :)