Sentiment learning using Twitter ideograms

It has been extremely important to know what the audience feels. With the growing technologies, people get to know what the majority thinks about a particular subject through review sites, people opinion voting sites, campaigns. Thus, it becomes necessary to get a conclusion on this so that the opinion-affected organization or individual gets a clear idea and works further. This survey introduces techniques and approaches that would directly help information-gathering systems. The present study focuses on methods that seek to address the new challenges raised by sentiment-aware applications, as compared to those that are already present in a more traditional-approach. Hence, we have tried to make the system more accurate by analyzing symbols which play a major role in sentiment learning. Proposed algorithm is compared and findings are presented in that explains each algorithm and its use in mining and analysis of Twitter textual data and provides deep insights as to what is the accuracy level in the mining process. We include a brief summary on broader issues such as privacy, manipulation and other related issues that opinion-access services gives rise to.