Introduction to News Analytics

News analysis is a technique in which various qualitative and quantitative attributes of text-based data (new stories) is measured to produce human understandable analytical conclusions. These results further help in manipulation of daily information in statistical format. The best use can be traced to financial markets. This analysis helps businesses and strategists to make better informed decisions. News analytics is the sub of text analysis.

Text analysis flow is as follows:

1. Data Mining
2. Data structuring
3. Algorithms analysis
4. Numeric Results

News analysis are usually derived through automated text analysis and applies to digital texts employing techniques from NLP (natural language processing) and machine learning such as latent semantic analysis, support vector machines, “bag of words” etc. The usage of these kind of algorithms have grown from an area of research to mature product solution since 2007. The sophisticated linguistic analysis of data from news and social media is always on a rise so is the research on the interpretation of that similar data into some format more usable and finding patterns.

The methods used for text analysis are:

• Word Frequency
• Collocation
• Concordance
• Text Classification
• Sentiment Analysis
• Topic Analysis
• Intent Detection
• Text Extraction
• Keyword Extraction
• Entity Extraction
• Word Sense Disambiguation
• Clustering

Why is News Analysis important?

According to MarketingProfs, 2 million articles on daily bases are produced and websites like worldometers.info track this info on live counter, simultaneously around 92000 articles are posted every 24 hours according to Chartbeat. That’s a colossal amount of data to process, and impossible for us to do it by ourselves, that is where we employ machines for sorting the data through data using text analysis models.

Other major factors are:

• Scalability
• Real-time Analysis
• Consistent Criteria

Where it is used?

New analytics and news sentiments calculations are being used many fields specifically in financial sector. The businesses use this to help them make better informed business decisions. Researchers are more involved in this due to their fascination by the results in predicting stock price movements, volatility and traded volume. Sentiments scores can be manipulated at various horizons to carter respective needs and objectives of frequency trading strategies.

Other applications include:

• Trading – Forex, Commodity, Equity etc.
• Business Decisions
• Academic Research
• Sentiments
• Algorithmic order execution
• Risk Management
• Strategizing
• Political campaigns

I know it is getting complicated while I have bombarded you with knowledge, leaving you to your feeling of overwhelmingness I will continue in next blog where I will try to show the real-time example of this and clarifying some of the terms used today.

Have any questions leave in the comments below I would love to revert.