Overview
Utilizing ML, Enin measures which news stories actually have predictive power, and to what degree. Despite sifting through the full breadth of available news and social media sources, they expose you to news which is both relevant and impactful. Enin analyzes the data from the perspective of risk, and displaying a unique approach to sentiment.Output sentiment is displayed in both raw & normalized format.
Data descriptions
Read_timestamp |
Timestamp |
Exchange_ticker |
Abbreviation of the exchange |
Instrument_ticker |
Ticker symbol for the security |
Instrument_name |
Full name of the security |
Exchange_name |
Full name of the exchange |
is_tradable |
A boolean indicating whether or not the instrument was directly tradable in the time period |
title |
The title part of the article text |
contents |
The main content of the article |
authors |
The authors of the article, if found |
combined_contents |
All text of the article, combined |
keywords |
Topic keywords that are detected |
summary |
NLP-based article summary |
positive_sentiment |
Positive sentiment, measured as number of positive words used in the article |
negatve_sentiment |
Negative sentiment, measured as number of negative words used in the article |
sum_sentiment |
Summation of the positive and negative sentiment |
normalized_positive_sentiment |
Positive sentiment divided by number of words in the article |
normalized_negative_sentiment |
Negative sentiment divided by number of words in the article |
normalized_sum_sentiment |
Sum sentiment divided by number of words in the article |
Benzinga’s data samples are intended to provide a data sample large enough for testing data quality and application for the financial markets. These sample files demonstrate a sample of the formats and content that can be delivered