iHealthcare: Predictive Model Analysis Concerning Big Data Applications for Interactive Healthcare Systems †

Abstract

Recently, the healthcare industry has caught the attention of researchers due to a need to develop a smart and interactive system for effective and efficient treatment facilities. The healthcare system consists of massive biological data (unstructured or semi-structured) which needs to be analyzed and processed for early disease detection. In this paper, we have designed a piece of healthcare technology which can deal with a patient’s past and present medical data including symptoms of a disease, emotional data, and genetic data. We have designed a probabilistic data acquisition scheme to analyze the medical data. This model contains a data warehouse with a two-way interaction between high-performance computing and cloud synchronization. Finally, we present a prediction scheme that is performed in the cloud server to predict disease in a patient. To complete this task, we used Random Forest, Support Vector Machine (SVM), C5.0, Naive Bayes, and Artificial Neural Networks for prediction analysis, and made a comparison between these algorithms.

A Genomic Information Management System for Maintaining Healthy Genomic States and Application of Genomic Big Data in Clinical Research

“…Currently, standards for clinical and genomic databases (DBs) should be drafted for an ideal maintenance strategy for the established CDW. Through data mining, disease factors can be discovered from the integrated DB, and a predictive model can be constructed [ 9 ]. Many methodologies or models that can discover insights from big data have been proposed.…”

Sentiment Analysis on Bengali Facebook Comments To Predict Fan's Emotions Towards a Celebrity

“…Thus, this score takes into account both false positives and false negatives. Although f1 is less intuitive than accuracy, it is generally more useful if the distribution of classes is uneven [28], [29]. When false positives and false negatives cost the same, accuracy works better.…”

“…Incorrect performance occurs when the system generates False Positives or False Negatives. For example, we want to predict emotion for a sentence which is actually 'happy' type [29]. If the machine predicts emotion as 'happy' then it will be True Positive.…”

“…It's not expected that a model will work correctly with real data it's never seen before. We need some assurance that the model has correctly identified the majority of the patterns in the dataset and is not overly sensitive to noise or has low bias and variance [29]. In order to get a static accuracy from SVM, we used cross-validation utilizing the K-fold algorithm, which allows for random sampling, and the stratified K-fold algorithm, which allows for stratified sampling.…”

Plant Disease Detection through the Implementation of Diversified and Modified Neural Network Algorithms

“…In [10], [11], [12], [13] and [14] the authors have applied Faster R-CNN with FCM-KM fusion for the rapid rice blast, bacterial blight, and blight disease detection. The authors have also used the 2DFM-AMMF algorithm for noise reduction and faster 2D-Otsu segmentation.…”