The problem of resolving references to earlier or later items in the discourse is commonly called as anaphora resolution or pronoun resolution. These items are usually noun phrases representing objects in the real world called referents but can also be verb phrases, whole sentences or paragraphs. Nowadays, anaphora resolution is addressed in numerous NLP (Natural Language Processing) applications. Proper treatment of anaphoric relations improves the performance of applications. Machine translation, information extraction, text summarization, or dialogue systems are some of the common applications of NLP. In early days, the machine translation systems processed on the basis of a sentence-by-sentence level. It did not consider the ties between sentences and resulted in an incoherent text as output. When the researcher forgets to handle the anaphora issue, it results in the striking problem of incorrect facts. It is very much needed to concentrate on the usage of pronoun, as it should match with their antecedents both in number and gender. Assigning inappropriate morphological features to the anaphor often may also lead to an undesirable change in the meaning of the sentence.

Sentiment analysis mainly focuses on subjectivity and polarity detection. Today consumer makes buying decision based on the customer's review that is available in each of the online shopping sites. There are some of the specific websites which discuss about positive and negative facts of those products that comes to market. Hence this type of analysis are socially very needed for sellers to undergo market analysis, branding, product penetration, market segmentation and so on. This paper mainly focuses on difference between systematic and automated methods of determining the positive and negative polarity distribution of Samsung Tablet PC.