Implement Machine Learning component V4 (dsmvp-v4)

As part of the progression of machine learning components with increasing levels of sophistication, implement version 4 ("dsmvp-v4") with the following characteristics:

Ensemble consensus learning: A machine learning module that can learn to detect racially biased expressions, assess the credibility of each marker and make use of them in computing the judgement for any input text. The labeled data will consist of triples.

A possible implementation of this claim may make use of what is known in the literature as "expert", which makes use of an ensemble of predictors (each corresponding to a model learned from a marker, or a group thereof), which based on errors made by each group, updates the ensemble weights multiplicatively so as to focus more on the more reliable markers in its learning process. (Reference: https://users.soe.ucsc.edu/~dph/mypubs/97experts.pdf)

This is to be combined with any of the other versions, dsmvp-v1 through v3.

Coding of dsmvp-v4 should be similar to and share many aspects of how dsmvp-v1 in the repository is implemented, using Jupyter notebook and accessing the database via webapi, etc.

See more details on a concrete algorithm for this functionality depicted in the attached ppt file.
[TakeTwo dsmvp v4 details.pptx](https://github.com/Call-for-Code-for-Racial-Justice/taketwo-datascience/files/5962394/TakeTwo.dsmvp.v4.details.pptx)