Definition
Sentiment analysis, also known as opinion mining is a subject area that uses computational techniques to analyze, process, and uncover people’s feelings, sentiments, and emotions embedded in a text or interaction. It involves the use of machine learning (ML), natural language processing (NLP), data mining, and artificial intelligence (AI) to mine, extract and categorize users’ opinions on a company, product, person, service, event, or idea for various sentiments [1].
For example, an AI reads a text from a customer of konga about their service;
Customer’s text:
"Konga delivered my phone to Lagos on time. Very impressed!".
The AI analysis
goes ahead to categorize it as positive sentiment with a score of 8.5 based on
the keywords "on time," and "impressed". It then takes the action of featuring the
review in marketing materials
Possible categories for scoring sentiments [2]
Origin
Sentiment analysis originated in the early 2000s when researchers began examining the automated extraction of sentiments from textual data as a way of quantifying public opinions and consumer perceptions. Throughout the years, sentiment analysis has achieved remarkable progress, driven by the proliferation of big data, social media, and the increasing importance of customer feedback in driving business strategies. The advancement of sentiment analysis has been marked by the merging of interdisciplinary fields such as linguistics, computer science, and cognitive psychology, fostering innovative methodologies for sentiment interpretation and sentiment-aware AI systems [3].
Context and
Usage
Sentiment analysis systems are used to analyze online sources such as emails, blog posts, online reviews, customer support tickets, news articles, survey responses, case studies, web chats, tweets, forums and comments. This helps companies and industries to gather insights into real-time customer sentiment, customer experience and brand reputation [4].
Why it Matters
Sentiment analysis assists companies to evaluate people's positive, negative, or neutral reactions to learn what people think. In the past, companies depended on traditional methods like surveys and focus groups to gather consumer feedback. But with machine learning and artificial intelligence technologies, it is now possible to analyze text from a variety of sources with greater accuracy and less effort. These analysis helps indicate if a product, service, or message needs to be adjusted to match an intended audience sentiment better [5].
In Practice
A good example
of a real-life case study of a company that offers sentiment analysis services
is Talkwalker. Their mission is to empower companies to make real time
decisions based on the true voice of their consumers. By listening to the
public conversations happening every minute across social and digital channels,
Talkwalker enables companies to understand what their consumers want.
See Also
Related NLP and Text Processing terms:
- Speech Analytics: Extracting insights and patterns from audio speech data
- Speech Recognition: Converting spoken language into written text
- Syntax Analysis: Understanding sentence structure
- Tokens: Individual units (words, subwords, characters) that text is divided into for processing
- Triple (Semantic Triple): Subject-predicate-object expressions
References
- Kanade, V. (2022). What is Sentiment Analysis? Definition, Tools, and Applications
- Getthematic. (n.d). Sentiment Analysis: Comprehensive Beginners Guide.
- Lark Editorial Team. (2023). Sentiment Analysis.
- Gillis, A., Barney, N. (2024). What is sentiment analysis?
- h2o. (2025). What is sentiment analysis?