The evolution of data science skill sets: An analysis using exponential family embeddings

Description

Many data scientists are familiar with word embedding models such as word2vec, which capture semantic similarity among words and phrases in a corpus. However, word embeddings are limited in their ability to interrogate a corpus alongside other context or over time. Moreover, word embedding models either need significant amounts of data or tuning through transfer learning of a domain-specific vocabulary that’s unique to most commercial applications.

Maryam Jahanshahi explores exponential family embeddings. Developed by Rudolph and Blei, these methods extend the idea of word embeddings to other types of high-dimensional data. Maryam demonstrates how this technique can be used to conduct advanced topic modeling on datasets that are medium sized, are specialized enough to require significant modifications of a word2vec model, and contain more general data types (including categorical, count, and continuous).

You’ll learn how TapRecruit implemented a dynamic embedding model using gensim and TensorFlow and its proprietary corpus of job descriptions. Using both categorical and natural language data associated with jobs, TapRecruit charted the development of different skill sets over the last three years. Maryam focuses on how data science and quantitative skill sets have developed, grown, and pollinated other types of jobs over time. If time allows, she’ll also discuss other segmentation analyses that were performed, including company types and geographies.