No abstract

Development of river water quality forecasting model (RWQFM) created using the concept of artificial neural network (ANN) for the river Ganga, India still has not been done as far as best awareness of the authors. In this research work an effort have been made for developing such model first time for the stream Ganga in the stretch from Devprayag to Roorkee, Uttarakhand, India by choosing five testing stations along this waterway. The month to month exploratory dataset for the time arrangement of 2001 to 2015 including four water quality parameters was taken. Using one of the proficient machine learning approach called ANN an optimal model is developed by conducting several experiments in Weka data mining tool. In advance the water quality is forecasted for next 12 months and the forecasting accuracy is determined using various performance measures. The computation of 12-steps ahead WQ indicated that the water comes out to be suitable for drinking throughout the year 2016 only at three stations: Devprayag, Rishikesh and Roorkee. At Haridwar station, the water is also comes out to be of best quality but only in nine months. In last quarter of 2016, a little degradation at Haridwar station while a crucial deterioration was noticed at Jwalapur site. The results showed that the proposed WQ model is more efficient in terms of the forecasting accuracy. At Rishikesh station the developed forecasting model achieved a noteworthy accuracy of 100%. Thus, the proposed ANN forecasting model is verified as an effective model and concluded that in overall the WQ of the Ganga River in this stretch is fine in 2016. Also, ANN has proven its significance as an efficient tool in the forecasting domain. Such models will definitely be helpful for the water management bodies in order to control the river pollution and consequently help the society as well

No abstract

DataStream mining is a challenging task for researchers because of the change in data distribution during classification, known as concept drift. Drift detection algorithms emphasize detecting the drift. The drift detection algorithm needs to be very sensitive to change in data distribution for detecting the maximum number of drifts in the data stream. But highly sensitive drift detectors lead to higher false-positive drift detections. This paper proposed a Drift Detection-based Adaptive Ensemble classifier for sentiment analysis and opinion mining, which uses these false-positive drift detections to benefit and minimize the negative impact of false-positive drift detection signals. The proposed method creates and adds a new classifier to the ensemble whenever a drift happens. A weighting mechanism is implemented, which provides weights to each classifier in the ensemble. The weight of the classifier decides the contribution of each classifier in the final classification results. The experiments are performed using different classification algorithms, and results are evaluated on the accuracy, precision, recall, and F1-measures. The proposed method is also compared with these state-of-the-art methods, OzaBaggingADWINClassifier, Accuracy Weighted Ensemble, Additive Expert Ensemble, Streaming Random Patches, and Adaptive Random Forest Classifier. The results show that the proposed method handles both true positive and false positive drifts efficiently.

No abstract