Sentiment Analysis from Text

This Bot takes text as input and extracts the global sentiment and polarity for the entities/keywords detected in the text.

Top Benefits

Detect if what is said in a comment or document is positive, negative, neutral, or has no polarity.
Opinion mining in news, social comments and contact center interactions.
Can perform granular analysis, assessing polarity both at a whole document level and at a sentence/segment/entity/concept
Identify subjectivity, irony and contradiction.
Easy to integrate and fully customizable.

Tasks

Analyze global sentiment expressed in a text
Extract entities in the text with the polarity associated to them
Extract concepts or keywords in the text with the polarity associated to them
Obtain full sentiment analysis of a text

This Bot carries out a request to MeaningCloud's Sentiment Analysis API. MeaningCloud's Sentiment Analysis does a complete morphosyntactic analysis and returns a complete sentiment analysis at a global, sentence and segment level. It also detects the entities (organizations, locations, people, etc.) and concepts (keywords) in the text, and the polarity associated with them to enable you to make aspect-based sentiment analysis.

Bot will output global polarity for the complete text, a list of the entities detected with their polarity in parentheses (and their type if configured), a similar list for concepts/keywords detected, and a JSON with the complete analysis returned by the API. Having the JSON response gives you access to all the information provided by the API, irony, subjectivity, as well as the sentences identified with the entities and concepts analysis at a sentence level.

Thanks to MeaningCloud, you can customize the analysis using user dictionaries (https://www.meaningcloud.com/developer/resources/dictionaries) to ensure the detection of the entities and concepts you want to analyze (and with the ontology type you want) and user sentiment models (https://www.meaningcloud.com/developer/resources/sentiment-models), to customize the sentiment analysis if the general scenario does not apply.
For instance, for the sentence The restaurant was great even though it’s not near Madrid., we will obtain a global sentiment analysis of P+ (Strong positive), the entity Madrid without any polarity (Madrid (NONE)) and the concept restaurant with strong positive polarity (restaurant (P+)).

Typical uses cases for Sentiment Analysis include social media analysis to analyze trends or brand reputation, Voice of the Customer in surveys or social media, etc.