Context is key
Problem being addressed
With the sheer amount of reviews and other opinions over the Internet, there is a need for automating the process of extracting relevant information. For machines, however, measuring sentiment is not an easy task, because natural language is highly ambiguous at all levels, and thus difficult to process.
A simple yet effective method for sentiment analysis using contextual embeddings and a self-attention mechanism. Leveraging contextual embeddings enables to convey a word meaning depending on the context it occurs in. The approach shows that attention-based models are suitable for other NLP tasks, such as learning distributed representations and sentiment analysis, and thus are able to improve the overall accuracy.
Advantages of this solution
The proposed sentiment classification model is language independent, which is especially useful for low-resource languages (e.g. Polish). The approach is comparable to the best performing sentiment classification models; and, importantly, in two cases yields significant improvements over the state of the art.
Solution originally applied in these industries
Possible New Application of the Work
With an explosion of digital information, social media interactions, and news highlights, it becomes cumbersome for banks to make informed decisions. NLP techniques overcome this limitation by classifying relevant news, articles, and comments to provide a real-time sentiment score. It enables banks to formulate well-informed investment strategies by extracting the risk signals out of market interactions.
Social sentiment analysis is the use of natural language processing to analyze social conversations online and determine deeper context as they apply to a topic, brand or theme. The net sentiment score and brand passion index show how users feel about a certain brand and how it compares across competitors, and using the context in this analysis will significantly improve its accuracy.
Source URL: #############