• Mr. Mithun M. Patil
• Mr. Vishwajit B. Gaikwad

Short textual data available online in the form of social media posts, product reviews, customer queries, search engine queries contain huge source of information, and extracting required knowledge out of it is valuable to business. However there are various challenges in analyzing such unstructured data. Shorts text such as social networking comments, customer reviews are usually grammatically incorrect, lacks sufficient statistical information to support many state-of-the-art approaches for text mining such as topic models. Further, they are more ambiguous with misspelled words and are generated in an enormous volume, which further increases the difficulty to handle them. In order to infer the actual meaning of short text it is essential to have semantic knowledge. After studying multiple methods proposed recently in the field of text analysis, this work has proposed a prototype system that uses sequential neural network for text processing. It divides the task into three subtasks as text segmentation, type detection and sentiment analysis. The proposed system has offline and online modules. In offline module, KERAS sequential model is build using twitter data to calculate affinity score and detect POS tags. Online model initially cleanse the data by filtering English stop words, followed by stemming using Snowball Stemmer and finally correct misspelled words using SpellChecker. The preprocessed data is then segmented using bi-gram and feed to KERAS sequential model to derive the semantic and sentimental knowledge of input text. Proposed method is tested on airline data and results are compared with some state-of-art methods. The analysis shows proposed method is better or equally effective compared to other methods.

POS tagging, bi-gram, KERAS Sequential model.

"Semantic and sentiment analysis of short text data", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 2, page no.d302-d306, February-2022, Available :http://www.jetir.org/papers/JETIR2202340.pdf

"Semantic and sentiment analysis of short text data", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 2, page no. ppd302-d306, February-2022, Available at : http://www.jetir.org/papers/JETIR2202340.pdf