Most if not all social media monitoring companies offer a mix of various technologies including sentiment analysis tools. However, numerous studies have shown that sentiment analysis is often not very accurate and also does not give a real insight into what is being said beyond the matching of key words against the text data.
If the key words change, or new language develops, or the style of grammar used is unusual, then sentiment analysis techniques can all have difficulties.
This is one of the reasons why Brand Aura focuses on contextual analysis. But what do we mean by this? Well, context for us means that the data drives the analysis (and not pre-selected key words). It means that each domain may have its own ‘language’ and yet still be analysed using the same process. It means that we find what concepts or words are in context with other concepts and words.
Since the data drives the output, it is easier for us to track changes between analysis runs (which could be done hourly, or daily, or weekly, depending on the amount of data). Since we find what is in context with key concepts, then we can identify more easily whether the comments are positive or negative (or otherwise). We can also delve much more deeply into the analysis and truly understand the main drivers behind the comments being made.