Method
fastText
Subword-based word vectors.
9 cases from 6 companies apply fastText. 1 was checked in an online A/B test; 1 report results in production.
Typical tasks: Evaluate models, Build a data platform, Make metric definitions consistent, Rank search results. Companies: Alfa-Bank, Airbnb, Rabota.ru, Zillow and 2 more. Often combined with: Word2Vec, Embedding similarity matching, BiLSTM, LSTM, Regular expressions and 33 more.
Machine Learning Service from Scratch, or Creating Frankenstein![]() | ![]() | general | 2024-11 |
| Airbnb Brandometer: Powering brand perception measurement on social media data with AI | travel | 2024-04 | |
Category Prediction for Search Query Understanding![]() | ![]() | e-commerce | 2024-04 |
| Salary Horrors and How to Avoid Them | classifieds | 2023-10 | |
| Development History of Voice Bot: Legacy Voice Bot of Alfa-Bank and Transition to Modern Methods | banking | 2023-05 | |
| NLP at Alfa-Bank, Part 3: Modeling | banking | 2023-05 | |
| NLP at Alfa-Bank, Part 9: Structuring OFD Receipts | banking | 2023-05 | |
| Incorporating Listing Descriptions into the Zestimate | classifieds | 2022-11 | |
| Neural network models on OFD receipt sequences | banking | 2022-06 |

