Sentiment Analysis using KNIME: a Systematic Literature Review of Big Data Logistics

Text analytics and sentiment analysis can help researchers to derive potentially valuable thematic and narrative insights from text-based content, such as industry reviews, leading operations management (OM) and operations research (OR) journal articles and government reports. The classification system described here analyses the aggregated opinions of the performance of various public and private, medical, manufacturing, service and retail organizations in integrating big data into their logistics. Although our results show a promising high level of model accuracy, we also suggest caution that the performance of the solution should be compared in terms of the performance of other solutions. This work explains methods of data collection and the sentiment analysis process for classifying big data logistics literature using KNIME (Konstanz Information Miner). Finally, it explores the potential of text mining to build more rigorous and unbiased models of operations management.