View article

Nowadays, the growth of the high-throughput technologies has resulted in exponential growth in the harvested data with respect to both dimensionality and sample size. The trend of this growth of the UCI machine learning repository is shown in Figure 1. Efficient and effective management of these data becomes increasing challenging. Traditionally manual management of these datasets to be impractical. Therefore, data mining and machine learning techniques were developed to automatically discover knowledge and recognize patterns from these data.

However, these collected data is usually associated with a high level of noise. There are many reasons causing noise in these data, among which imperfection in the technologies that collected the data and the source of the data itself are two major reasons. For example, in the medical images domain, any deficiency in the imaging device will be reflected as noise for the later process. This kind of noise is caused by the device itself. The development of social media changes the role of online users from traditional content consumers to both content creators and consumers. The quality of social media data varies from excellent data to spam or abuse content by nature. Meanwhile, social media data is usually informally written and suffer from grammatical mistakes, misspelling, and improper punctuation. Undoubtedly, extracting useful knowledge and patterns from such huge and noisy data is a challenging task.