Sentiment analysis for mixed script Indic sentences

Sentiment Analysis on Indian Indigenous Languages: A Review on Multilingual Opinion Mining

“…It helped in increasing the accuracy of the system because if the word was not present in the Hindi SWN then it found the closest word and assigned the score of that word [13]. In the study, Bhargava et al [14] completed the SA task on the FIRE 2015 dataset. The dataset consisted of code-mixed sentences in English along with 4 Indian languages (Hindi, Bengali, Tamil, Telugu).…”

“…SWN's of each language were used for sentiment classification. The results of the implemented system were compared with the previous language translation technique and 8% better precision was observed [14].…”

“…[8,9,12,13,15,23]. Authors in [14] have accessed the FIRE 2015 dataset. The dataset has 792 utterances and has 8 different languages other than English.…”

“…Also, suggestions to compare the algorithms and find the best algorithm was given by [6,7,18]. In [14] Machine learning algorithms LR and SVM were used for Language detection and SWN was used for classification of sentiments. SVM outperformed LR in Language detection.…”

“…Neutral polarity was not considered which could affect the analysis to greater extent. In [14] transliteration of words caused issues. Authors in [15] faced issue in automatic detection of the topic hashtags because the context was no provided to the system.…”