REFERENCES
Building a prescriptive model for advertisement text (with revenue uplift 7%) – conducted feasibility study for user recommendations based on recurrent neural networks.
Python, TensorFlow, RNN, Machine & Deep Learning
Text Mining: implementation of a new, scalable item based collaborative filtering algorithm for 150 million events per day.
Increase in click through rate by 10% compared to previous implementation – constructed real time streaming pipeline to transform newspaper text into vector format for about 20 texts per second being transformed, inserted and updated each day, implemented a fully automated model retraining pipeline.
Python, Spark, TensorFlow, Machine & Deep Learning, Big Data
Design and implement a Data Science Stack for Text Mining.
Regression, Classification and Apache Spark, Recurrent Neural Networks, Keras, Hadoop
Topic modeling: Use LDA-Algorithm to model the topics of German news
Doc2Vec: Use the topic model to transform each document to a topic vector
Similar documents detection: Use the topic model to transform the corpus to a topic matrix and search similar documents using the corpus-topic matrix
Keyword identification: Build a keyword extractor to identify keywords from German news
Python, Gitlab, Jmeter, SonarQube, Docker, Rancher
Analyze the text of product names, descriptions, etc. to improve search quality, develop novel POS and entity tagger and build a prediction model of the product categories by training a feedforward neural network.
Python, NLTK, Sklearn, machine learning, Neural Networks
Web-mining & Text-analytics: crawl websites, apply text analytics techniques to extract information.
Natural language processing for German: POS-tagging and stemming based on statistical inference; topic and sentiment analysis of the news.
R, R-Shiny, IBM Watson
© 2021, lexoro GmbH
