Sanchari Chatterjee1, Angelina A. Tzacheva2, and Zbigniew Ras1, 1Department of Computer Science, University of North Carolina at Charlotte, Charlotte, NC, 28223, USA, 2Computer Science and Information Technology, College of Computing and engineering, WestCliff University, Irvine, CA 92614
Education sector, Business field ,Medical domain and Social Media, huge amounts of data in a single day . Mining this data can provide a lot of meaningful insights on how to improve user experience in social media, users engage in these domains collect and cherish the data as they hope to find patterns and trends and the golden nuggets that help them to accomplish their goal. For example: How to improve student learning; how to increase business profitability; how to improve user experience in social media; and how to heal patients and assists hospital administrators. Action Rule Mining mines actionable patterns which are hidden in various datasets. Action Rules provide actionable suggestions on how to change the state of an object from an existing state to a desired state for the benefit of the user. There are two major frameworks in the literature of Action Rule mining namely Rule-Based method where the extraction of Action Rules is dependent on the pre-processing step of classification rule discovery and Object-Based method where it extracts the Action Rules directly from the database without the use of classification rules. Hybrid Action rule mining approach combines both these frameworks and generates complete set of Action Rules. The hybrid approach shows significant improvement in terms computational performance over the Rule-Based and Object-Based approach. In this work we propose a novel Modified Hybrid Action rule method with Partition Threshold Rho, which further improves the computational performance with large datasets.
Actionable Patterns, Action Rules, Emotion Detection, Data Mining, Rule-Based, ObjectBased.
Escalona González Sergio1, González Milán Zoraida2, Ricardo Paez Beatriz3, 1Department of Nephrology, University of Medical Sciences of Las Tunas, Las Tunas, Cuba, 2Department of Nephrology, General Teaching Hospital: "Dr. Ernesto Guevara de la Serna", Las Tunas, Cuba, 3Department of Nephrology, University of Medical Sciences of Las Tunas, Las Tunas, Cuba
Artificial neural networks are a promising field of artificial intelligence in disease and mortality prediction. A retrospective cohort study was conducted in 54 patients. The training and testing of a multilayer neural network with architecture for the prediction of mortality from chronic kidney disease was carried out, the efficiency of the neural network was evaluated by means of the area under the curve of the C statistic. Finally, the importance of the variables was analyzed. The training of the artificial neural network showed 81.0% accuracy for survival prediction, 93.8% for mortality and 86.5% overall. The test showed 100% accuracy for survival prediction, 81.8% for mortality, and 88.2% overall. The area under the curve was of 0.936. An artificial neural network was implemented for the prediction of mortality due to chronic kidney disease with optimal statistical indicators.
Chronic Kidney Disease, Mortality, Prediction, Artificial Intelligence, Artificial Neural Networks.
Sainah Sreeram and Ananya Tyagi, Vivekananda Institute of Professional Studies, Pitampura, New Delhi
Sentiment analysis is a method used in natural language processing (NLP) to identify the emotional component of a text. It is sometimes alluded to as opinion mining. It has been widely used around the world to improve service outcomes and customer experiences, as it further increases client trust and management. However, there are still unresolved issues in the relatively new field of opinion mining research. Depending on the domain of the dataset, some issues are common to opinion mining in general, while others are exclusive to their own sources and context. A thorough description of these sentiment analysis concerns is provided in this paper along with a detailed analysis of various classifiers at different levels. Later, this paper also emphasised text summarisation, requirement engineering as well as ensemble methods on twitter sentiment analysis.
sentiment analysis, natural language processing.
