NetOwl Extractor offers advanced sentiment analysis capabilities by leveraging its industry-leading entity extraction along with its intelligent natural language processing (NLP) technologies. NetOwl’s sentiment analysis goes beyond traditional sentiment analysis where “positive” or “negative” sentiment is assigned at the document or sentence level. Such traditional approaches fail to recognize multiple, sometimes conflicting, sentiments that exist within a single document or sentence. They also fail to capture the precise object of a sentiment.
Entity- and Aspect-based Sentiment Analysis
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 provides finer-grained sentiment analysis directed toward different aspects of entities, for example, the price of a product, the new policy of a country, the campaign of a presidential candidate, etc. In this way, NetOwl pinpoints what the sentiment is and precisely what the sentiment is about, making sentiment analysis much more informative and useful to its users
- Allows much more detailed sentiment analytics on each entity
- Addresses cases where multiple, conflicting sentiments are expressed within a single document or sentence
Sophisticated Sentiment Ontology
NetOwl’s sentiment analysis 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, our 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 an organization to prioritize and tailor its responses in a much more targeted fashion.
- Allows more in-depth analysis of opinions and actions
- Enables organizations to offer more appropriate, prioritized responses
Semantic DisambiguationRecognizes and classifies concepts using linguistic context. This sophisticated feature distinguishes semantic ambiguities like:
- "Apple" (company) vs. "apple" (fruit)
- "hip" hotel vs. "hip" replacement
- "popular" movie vs. "popular" vote
Language IDOffers a seamlessly integrated language ID capability where the language of the input text is automatically detected, and the text is processed accordingly. Both microblog and standard document lengths are supported. A mixed language document, where sections of the document are written in multiple languages, can also be handled automatically.
Name NormalizationAssigns normalized forms to extracted person, organization, and place names, taking into account capitalization, acronyms, abbreviations, nicknames, etc. When NetOwl’s Smart Geotagging is used alongside entity extraction, place names are both disambiguated and normalized. Name normalization is ideal for cross-document name resolution for various social media analytics applications.
Coreference ResolutionNetOwl Extractor resolves co-referring extracted entities, whether they are name aliases, pronouns, or definite noun phrases, identifying them as referring to the same object. For example:
- "UAL" ➞ "United Airlines"
- "The company’s founder" ➞ "Steve Jobs"
Social Media AnalyticsNetOwl Extractor has been trained on extensive data from various social media platforms, including Twitter, Facebook, and blogs.
MultilingualSentiment Analysis is currently available in English and Arabic. Other languages are on the way.