SenticNet is a Singapore-based company offering state-of-the-art sentiment analysis tools for tasks such as brand positioning, trend discovery, and social media marketing. Unlike statistical approaches to opinion mining, SenticNet focuses on a semantic-preserving representation of natural language concepts and on sentence structure. In particular, SenticNet overcomes common limitations of COTS sentiment analysis tools by proposing the following three shifts:
1. Shift from mono- to multi-disciplinarity
evidenced by the concomitant use of AI and Semantic Web techniques, for knowledge representation and reasoning; mathematics, for tasks such as graph mining and multi-dimensionality reduction; linguistics, for discourse analysis and pragmatics; psychology, for cognitive and affective modeling; sociology, for understanding social network dynamics and social influence; finally ethics, for understanding related issues about the nature of mind and the creation of emotional machines. [Read more]
2. Shift from syntax to semantics
enabled by the adoption of the bag-of-concepts model in stead of simply counting word co-occurrence frequencies in text. Working at concept-level entails preserving the meaning carried by multi-word expressions such as cloud_computing, which represent ‘semantic atoms’ that should never be broken down into single words. In the bag-of-words model, the concept cloud_computing would be split into computing and cloud, which may wrongly activate concepts related to the weather. [Read more]
3. Shift from statistics to linguistics
implemented by allowing sentiments to flow from concept to concept based on the dependency relation between clauses. The sentence “iPhone6 is expensive but nice”, for example, is equal to “iPhone6 is nice but expensive” from a bag-of-words perspective. However, the two sentences bear opposite polarity: the former is positive as the user seems to be willing to make the effort to buy the product despite its high price, the latter is negative as the user complains about the price of iPhone6. [Read more]
We all have preferences about the products and services we use. We also like to gossip about actors, sportsmen, and politicians. Whenever we find the right time, mood, or context to share such feelings, we start the opinon sharing process.
In order for opinions to be shared, they need to be 'encoded' in natural language and through a Web format. This is when problems start: opinions on the Web are meant for human-consumption and, hence, not directly machine-intelligible.
By leveraging on the ensemble application of common-sense computing, machine learning and linguistics, SenticNet can bridge the gap between unstructured social data and structured machine-processable statistics like no-one else.
Experience SenticNet for the basic sentiment analysis tasks of polarity detection and emotion recognition from short-text messages
Less limitations on API calls, but you can get a real taste of SenticNet for advanced opinion mining tasks such as aspect-based sentiment analysis
Unleash the full power of Sentic Patterns for tasks such as brand positioning, trend discovery, social media marketing, and more