Sentimental Analysis

Sentiment analysis is the interpretation and classification of emotions within text data using text analysis techniques. It bifurcates the data into positive, negative and neutral. Sometimes, there are more emotions that also can be detected using sentimental analysis.
1583480-online-education-png-education-png-456_413_preview
This analysis tools allow businesses and other decision makers to identify customer  sentiment toward products, brands, campaigns or services in online feedback. It is highly used to check online users attitude using their tweet, texts, reviews and comments. According to Wikipedia, Sentiment analysis refers to the use of natural language processing, text analysis, computational linguistics, and bio-metrics to systematically identify, extract, quantify, and study effective states and subjective information. It is a Natural Language Processing and Information Extraction task that aims to obtain writer’s feelings expressed in positive or negative comments, questions and requests, by analyzing a large numbers of documents. 
Uses:
  • Finance field for stock trading companies as sentiment algorithms can detect particular companies who show a positive sentiment in news articles. This can mean a significant financial opportunity, as this may trigger people to buy more of the company’s stock. Having access to this type of data, may give traders the opportunity to make decisions before the market has time to react. (getthematic)
  • Social media posts to know how are customer reacting to a particular topic, concept, reviews and what are their sentiments for the same.
  • Increase customer retention.
  • New product perception.
  • Businesses gain valuable insights using sentimental analysis and can strategize their objectives according to the output of this technique.
  • Also provides company’s to gain competitive advantage.

Sentiment-Analysis

Approach:

  • Using NLP (Natural Language Processing) that allows computers to analyze and understand human language.
  • Using Machine learning concept that uses classifiers such as Naive Baiyes, SVM etc.

 

Working:

  • Collection/ Fetching of data: A large chunk of data is collected for further analysis.
  • Cleaning the data: Since the data collected is unstructured or semi-structured in nature, it has to be cleaned before going for analysis. It is done by removing stop-words, removing punctuation, whitespaces etc.
  • Data Analysis: The inbuilt algorithm process the data and perform sentence splitting.
  • Indexing: Algorithm tags sentences based on their polarity such as positive, negative or neutral.
  • Output: Finally, it provides an output that tells the actual results. It can be in the form of table, graphs, text, word-cloud or figures.

 

Limitations:

  • Latest emerging terms cannot be analysed properly.
  • Sarcasm detection.
  • Term frequency.
  • Sentence completeness.
  • Small sentences or phrases cannot give much insights.