Text as Alternative Data: An Introduction with Forecasting Applications, May 8-19

Get to know how text mining is used in larger pipelines that provide concrete decision-making support

This 20h online course will enable you to integrate textual data into your work environment after only two weeks of training. You will be able to use dictionary-based sentiment analysis for stock market evaluations, use the topic model to conduct political risk predictions, and use BERT to conduct sentiment analysis on financial text.

After successful completion of the course, you will:

• Understand text pre-processing
  Text pre-processing is an essential step in the text mining process, as it prepares the text data for further analysis. By learning text pre-processing techniques such as tokenization and stemming, you will be able to effectively clean and prepare text data for analysis.

• Have learnt LDA
  Latent Dirichlet Allocation (LDA) is a popular technique for topic modeling that allows you to identify the main themes or topics present in a collection of documents. This can be useful for text summarization, document classification, and information retrieval.

• Understand sentiment analysis
  Sentiment analysis is the process of identifying the sentiment expressed in a piece of text, whether it be positive, negative, or neutral. By learning sentiment analysis techniques, you will be able to extract valuable insights and information from large amounts of text data.

• Have learnt BERT
  BERT is a state-of-the-art transformer-based language model that has achieved impressive results on a variety of natural language processing tasks. By learning about BERT, you will be able to use this powerful tool to perform tasks such as text classification and language translation.

Get the tools to effectively analyze and extract insights from large amounts of text data, a valuable skill in today's data-driven world

This course is specifically aimed at: 

Professionals in all sectors (finance, consultancy, industry, public sector, international organizations) are in contact with large amounts of documents/text but don’t know how to access the information in these documents in an automated way.

People that want to predict outcomes or track sentiment towards a product or stock and have access to text sources that can be matched to these targets.

The target participant must have knowledge of Python and machine learning but can be from any academic or working background.

You can find the full course outline and more details on the course page