Word Embeddings (DELTA)
-
- 0 Rating
- 0 Reviews
- 5 Students Enrolled
Word Embeddings (DELTA)
Word embeddings map words to dense vectors that reflect meaning based on contextual usage, improving over one-hot encodings. Popular methods include Word2Vec (CBOW/Skip-gram with negative sampling), with GloVe leveraging global co-occurrences and FastText adding subword n-grams for morphology and rare words.
-
- 0 Rating
- 0 Reviews
- 5 Students Enrolled
Requirements
- see DELTA course requirements
General Overview
Description
Introduction to word embeddings: Dense vector representations of words designed to capture semantic and syntactic relations in a continuous space. Desiderata include proximity for similar words, smooth generalization, and arithmetic regularities; contrasted with one-hot vectors, which are sparse and encode no similarity. Rooted in the distributional hypothesis—words occurring in similar contexts tend to have similar meanings.
The Word2Vec (W2V) algorithm: Learns embeddings via self-supervised training with a shallow neural network using CBOW (predict a word from its context) or Skip-gram (predict context from a word), typically optimized with negative sampling or hierarchical softmax.
Beyond W2V—alternative strategies: GloVe derives embeddings from global word–word co-occurrence statistics via a weighted least-squares objective, while FastText enriches representations with character n-grams to model morphology and handle rare or out-of-vocabulary words.
Courses that include this CL
Recommended for you
Meet the instructors !
Stefan received a PhD from the University of Hamburg in 2007, where he also completed his habilitation on decision analysis and support using ensemble forecasting models in 2012. He then joined the Humboldt-University of Berlin in 2014, where he heads the Chair of Information Systems at the School of Business and Economics. He serves as an associate editor for the International Journal of Business Analytics, Digital Finance, and the International Journal of Forecasting, and as department editor of Business and Information System Engineering (BISE). Stefan has secured substantial amounts of research funding and published several papers in leading international journals and conferences. His research concerns the support of managerial decision-making using quantitative empirical methods. He specializes in applications of (deep) machine learning techniques in the broad scope of marketing and risk analytics. Stefan actively participates in knowledge transfer and consulting projects with industry partners; from start-up companies to global players and not-for-profit organizations.