We just released a new update for the EXIOBASE 3 MRIO time-series of monetary tables: v3.8.
It is a full re-estimate of the time-series, but still relies heavily on "now-casting" economic structure (and environmental extensions).
Google Trends show you the search-term frequency of a specific term relative to the total search-volume.
As by now, their is no official API interface to Google trends. Their are, however, some unofficial packages to access the Google Trends. One of them is pytrends which we will use here to get the relaitve search-term frequency.
Sentiment analysis estimates whether whether a piece of text is negative, neutral or positive. There are two approaches towards sentiment analysis: binary or polarity based or intensity/valence based. The polarity based only provides information if a certain text is postive or negative. For example, 'good' and 'perfect' would score the same. In contrast, a valence based sentiment analysis also takes the intensity of a word into account, therefore giving a higher value to 'perfect' in the example above.
One of the major applications of sentiment analysis is judge social media streams (Twitter, facebook posts). For example, companies use it to identify negative feedbacks on social media and react/answer to them.
This notebook explains how to search the live twitter stream for certain hashtags and keywords. The gathered tweets are stored as text files and further processed into a panda DataFrame.
The script is designed for long term streaming with gathered tweets stored in succeeding files.
Traditionally, the initial blog post on a custom made site desribes how to setup and build the site. However, there are already a plethora of such descriptions available. Therefore, I here rather share my motivation for the why, the outcome of my research about the how, followed by some points about practicalities.