Sentiment Analysis icon

Sentiment Analysis

Entity-based Sentiment Analysis goes far beyond traditional sentiment analysis where “positive” or “negative” sentiment is assigned at the document or sentence level.  NetOwl recognizes the multiple, sometimes conflicting, sentiments about entities that may exist within a single document.

Abstract illustration of points on a chart showing spikes and valleys in data
Icons of different facial expressions

Entity- and Aspect-based Sentiment Analysis

Simple approaches to sentiment analysis that classify a text or a sentence as positive, negative, or neutral are of limited value because they can’t capture what the sentiment is actually about.

NetOwl offers entity-based sentiment analysis as well as aspect-based (or feature-based) sentiment analysis. At the entity level, NetOwl identifies sentiments toward various types of entities such as people, organizations, brands, and products. At the aspect level, NetOwl captures the specific entity aspects that sentiments are about, for example, the price of a product, the new policy of a country, the campaign of a presidential candidate, etc.

By being able to pinpoint what the sentiment is about, NetOwl’s sentiment analysis is much more informative and useful because it:

  • Allows much more detailed sentiment analysis on each entity (e.g., “My iPad speaker does not work.”)
  • Captures multiple, conflicting sentiments expressed within a single document or sentence. (e.g., “I really love my car, but its gas mileage is disappointing.”)

Key Product Features

Arrow in target icon

Accurate

Provides state-of-the-art sentiment analysis accuracy even on noisy text. 

Extensive Entity Coverage

Handles all types of entities as targets of sentiment – named and non-named entities of major entity types and beyond.

Overlapping chat bubbles icon

Semantic Disambiguation

Distinguishes semantic ambiguities such as “cool” in “cool wind” vs. “cool car” by utilizing linguistic context.

Browser with highlighted data icon

Coreference Resolution

Resolves co-referring entities, whether they are names, pronouns, definite noun phrases, or events: “the technology company” → “Apple”.

Scalable icon

Fast & Scalable

Extremely fast for real-time analysis. Highly scalable sentiment analysis software with Docker and Kubernetes support.

Computer API icon

Easy Integration

Easy-to-integrate entity-based sentiment analysis product with a REST API.

illustration representing entity extraction

Refined Sentiment Ontology

NetOwl’s sentiment analysis tool does not just give “negative” or “positive” binary answers, but offers a more refined sentiment ontology to distinguish different opinions, attitudes, intentions, and behaviors. For example, NetOwl’s fine-grained sentiment ontology makes a distinction between a complaint about a product feature and a threat to boycott a brand.

This kind of distinction will allow organizations to conduct a more in-depth sentiment analysis of opinions and actions. In addition, organizations are able to prioritize and tailor their responses in a much more targeted fashion.

Normalization for Trend and Predictive Analysis

NetOwl’s sentiment analysis software assigns normalized forms to extracted entities (both names and descriptive phrases) and sentiment expressions. It takes into account capitalization, acronyms, abbreviations, nicknames, morphological variants (e.g., number, tense), etc.

This normalization capability is critical to be able to classify, aggregate, and quantify the myriad sentiments expressed in a large data set. NetOwl allows production of much more accurate and meaningful graphs, charts, and dashboards for more effective sentiment analysis, discovery, and monitoring.

A dashboard interface with a header bar, vertical and horizontal axes with arrows, teal-colored bar.
Illustration showing sentiment analysis applications

Sentiment Analysis Applications

NetOwl’s sentiment analysis has been trained extensively on data from various social media platforms, including Twitter, Facebook, and blogs over many domains. It is useful for many applications, including CRM, online reputation monitoring, market research, online marketing, employee hiring and retention, online review monitoring, election campaigns, public opinion, and geopolitical monitoring.

Sentiment Analysis Solutions

Social media bubbles with numbers of likes

Social Media Analysis

Social media is a gold mine of opinion data, but extracting actionable insights from this unstructured mass remains a major challenge.

Media screens on a dark background

Adverse Media Monitoring

Mitigate corporate risk by instantly identifying negative coverage across your supply chain and client base.

Globe wrapped in data

Intelligence Analysis

With 80% of intelligence data being unstructured, analysts need scalable text analytics to unlock vital national security insights.

Trusted by leading global organizations

LexisNexis Logo
Dun & Bradstreet logo
Delta Dental logo
RBC logo
Deluxe logo
Light gray diagonal shape spanning the full width of the canvas.

Featured Blog Posts

Sentiment Analysis

What is Entity-Based Sentiment Analysis?

There is so much opinion data out there, but how do we know what people are actually saying?

View Post

How to choose a Sentiment Analysis product

How to Choose a Sentiment Analysis Product

Sentiment Analysis can be an effective tool to better understand your corporate reputation or how your products are being received.

View Post

Sentiment Analysis for brand monitoring

Sentiment Analysis Provides Large-Scale, Real-Time Brand Monitoring

In a global and fast-paced economy, monitoring the many aspects that are part of a company’s brand reputation is more…

View Post

Frequently Asked Questions

  • How is entity-based Sentiment Analysis more useful than other types of Sentiment Analysis?

    Some common approaches to Sentiment Analysis such as traditional machine learning-based approaches are unable to reliably determine what a sentiment is about. They also fall short when there are multiple possibly conflicting sentiments expressed in the same sentence or document.

    By contrast, entity-based Sentiment Analysis is able to identify the specific entities (e.g., people, organizations, brands, and products) and specific aspects (e.g., the price of a product, the new policy of a country) that sentiments are associated with, allowing for a more detailed sentiment analysis on each entity, even when there are conflicting sentiments expressed within a single sentence or document.

  • What is Sentiment Analysis used for?

    Sentiment Analysis supports many critical applications such as Social Media Analysis, Reputation Management, Brand Monitoring, Market Research, CRM, Voice of the Customer, Competitive Analysis, Public Opinion Mining, Marketing Campaigns Analysis, Political Campaign Analysis, Voice of the Employee, etc. It is widely used in both the commercial world and Government.

  • Can Adverse Event Extraction supplement Sentiment Analysis?

    Yes. NetOwl’s unique Adverse Event Extraction offered through its Event Extraction capability can provide important additional information about entities of interest mentioned in unstructured text.

    Negative events captured by Adverse Event Extraction (e.g., arrests, bankruptcies, indictments, product recalls) may affect an entity’s or brand’s reputation and risk profile.