Nikita Zhiltsov
Analyst @ Rambler & Co
Moscow

Modern means of NLP in search tasks

  • Auditorium: ARCHIVE
  • Category: Development
  • Level: Skilled
  • Section: Machine learning

Description:

The report is devoted to modern text processing facilities based on machine learning, applied for some search tasks in Rambler & Co projects (portal, LJ). The speaker shares his experience in developing solutions based on the vector representation of word2vec and neural networks learning on real data. Examples of using the libraries of fastText, Keras and Tensorflow will be considered.

О спикере

Nikita Zhiltsov

Analyst @ Rambler & Co
Moscow

Carry out R&D in search sphere and natural language processing. Published more than 16 scientific papers, including publications at such leading conferences in the field of information search as ACM SIGIR and CIKM.

In 2014 co-founded the Textocat company, which develops solutions in the field of intellectual analysis of texts and chat bots.

In 2016 defended his Candidate thesis on technical sciences "Methods and algorithms of searching for entities by key words in graph knowledge bases."

Since 2016 - data analyst in the team of search technologies at Rambler & Co.

Похожие доклады