Adnan Idris

Churn prediction in telecom using Random Forest and PSO based data balancing in combination with various feature selection strategies

The telecommunication industry faces fierce competition to retain customers, and therefore requires an efficient churn prediction model to monitor the customers churn. Enormous size, high dimensionality and imbalanced nature of telecommunication datasets are main hurdles in attaining the desired performance for churn prediction. In this study, we investigate the significance of a Particle Swarm Optimization (PSO) based undersampling method to handle the imbalance data distribution in collaboration with different feature reduction techniques such as Principle Component Analysis (PCA), Fishers ratio, F-score and Minimum Redundancy and Maximum Relevance (mRMR). Whereas Random Forest (RF) and K Nearest Neighbour (KNN) classifiers are employed to evaluate the performance on optimally sampled and reduced features dataset. Prediction performance is evaluated using sensitivity, specificity and Area under the curve (AUC) based measures. Finally, it is observed through simulations that our proposed approach based on PSO, mRMR, and RF termed as Chr-PmRF, performs quite well for predicting churners and therefore can be beneficial for highly competitive telecommunication industry.

Intelligent churn prediction in telecom: employing mRMR feature selection and RotBoost based ensemble classification

Churn prediction in telecom has recently gained substantial interest of stakeholders because of associated revenue losses.Predicting telecom churners, is a challenging problem due to the enormous nature of the telecom datasets. In this regard, we propose an intelligent churn prediction system for telecom by employing efficient feature extraction technique and ensemble method. We have used Random Forest, Rotation Forest, RotBoost and DECORATE ensembles in combination with minimum redundancy and maximum relevance (mRMR), Fisher’s ratio and F-score methods to model the telecom churn prediction problem. We have observed that mRMR method returns most explanatory features compared to Fisher’s ratio and F-score, which significantly reduces the computations and help ensembles in attaining improved performance. In comparison to Random Forest, Rotation Forest and DECORATE, RotBoost in combination with mRMR features attains better prediction performance on the standard telecom datasets. The better performance of RotBoost ensemble is largely attributed to the rotation of feature space, which enables the base classifier to learn different aspects of the churners and non-churners. Moreover, the Adaboosting process in RotBoost also contributes in achieving higher prediction accuracy by handling hard instances. The performance evaluation is conducted on standard telecom datasets using AUC, sensitivity and specificity based measures. Simulation results reveal that the proposed approach based on RotBoost in combination with mRMR features (CP-MRB) is effective in handling high dimensionality of the telecom datasets. CP-MRB offers higher accuracy in predicting churners and thus is quite prospective in modeling the challenging problems of customer churn prediction in telecom.

Wavelet Based Image Authentication and Recovery

In this paper, we propose a secure semi-fragile watermarking technique based on integer wavelet transform with a choice of two watermarks to be embedded. A self-recovering algorithm is employed, that hides the image digest into some wavelet subbands for detecting possible illicit object manipulation undergone in the image. The semi-fragility makes the scheme tolerant against JPEG lossy compression with the quality factor as low as 70%, and locates the tampered area accurately. In addition, the system ensures more security because the embedded watermarks are protected with private keys. The computational complexity is reduced by using parameterized integer wavelet transform. Experimental results show that the proposed scheme guarantees safety of a watermark, recovery of image and localization of tampered area.

Genetic Programming and Adaboosting based churn prediction for Telecom

Churn prediction model guides the customer relationship management to retain the customers who are expected to quit. In recent times, a number of tree based ensemble classifiers are used to model the churn prediction in telecom. These models predict the churners quite satisfactorily; however, there is a considerable margin of improvement. In telecom, the enormous size, imbalanced nature, and high dimensionality of the training dataset mainly cause the classification algorithms to suffer in accurately predicting the churners. In this paper, we use Genetic Programming (GP) based approach for modeling the challenging problem of churn prediction in telecom. Adaboost style boosting is used to evolve a number of programs per class. Finally, the predictions are made with the resulting programs using the higher output, from a weighted sum of the outputs of programs per class. The prediction accuracy is evaluated using 10 fold cross validation on standard telecom datasets and a 0.89 score of area under the curve is observed. We hope that such an efficient churn prediction approach might be significantly beneficial for the competitive telecom industry.

Short Term Earthquake Prediction in Hindukush Region Using Tree Based Ensemble Learning

Earthquake prediction has been long considered as impossible phenomenon but recent research studies show some progress in this field by considering it as a data mining problem. There are numerous challenges in earthquake prediction, which includes highly non-linear behavior of seismic activity and non-availability of reliable seismic precursors. This work focuses on earthquake prediction in Hindukush region by employing mathematically computed seismic features and using these features to model earthquake occurrences through employing machine learning techniques. The study aims to consider earthquake prediction as a binary classification problem. The short term earthquake prediction is performed using tree based ensemble classifiers, where rotation forest has shown good prediction results, compared to random forest and rotboost.

Churn Prediction System for Telecom using Filter–Wrapper and Ensemble Classification

Churn prediction in telecom is a challenging data mining task for retaining customers, especially, when we have imbalanced class distribution, high dimensionality and large number of samples in training set. To cope with this challenging task of churn prediction, we propose a new intelligent churn prediction system for telecom, named FW-ECP. The novelty of the FW-ECP lies in its ability to combine both filterand wrapper-based feature selection as well as exploit the learning capability of an ensemble classifier built using diverse base classifiers. In the filter phase, Particle Swarm Optimization-based undersampling and mRMR feature selection are employed to reduce the effect of imbalanced class distribution and large dimensionality. In Wrapper phase, we employ Genetic Algorithm that further discards irrelevant and redundant features. Random Forest, Rotation Forest, RotBoost and SVMs are then employed to exploit the new feature space. Finally, the ensemble classifier is constructed using both majority voting and stacking. We have tested and compared the performance of proposed FW-ECP system on two publicly available standard telecom datasets: Orange and Cell2Cell. FW-ECP takes into account both the imbalanced nature and large dimensionality of the training sets and yields better prediction performances compared with existing state-of-the-art approaches. The feature spaces for the Orange and Cell2Cell datasets are reduced to 24D and 18D, from 260D and 76D, respectively. The AUCs obtained by FW-ECP are 0.85 and 0.82 for Orange and Cell2Cell datasets, respectively.

Ensemble Based Efficient Churn Prediction Model for Telecom

Predicting churners in telecom is an important application area of pattern recognition that helps in responding appropriately for retaining customers and saving the revenue loss a corporation suffers. The aim of the churn predictor is to learn the pattern of churners and thus differentiate between churners and non-churners. Handling the large dimensionality and selecting discriminative features are challenging aspects of telecom churn prediction that hinder the performance of predictors. In this study, we propose a churn prediction approach that exploits the discriminative feature selection capabilities of minimum redundancy and maximum relevance in the first step, leading to enhanced feature-label association and reduced feature set. The diverse ensemble of different base classifiers is then applied as a predictor in a second step. Final predictions are computed based on majority voting Random Forest, Rotation Forest and KNN, that ultimately leads to predicting churners from telecom datasets with higher accuracy. Simulation results are evaluated using sensitivity, specificity, area under the curve (AUC) and Q-statistic based measures on standard telecom datasets. The results indicate that our proposed approach efficiently models the challenging problem of telecom churn prediction, by effectively handling the large dimensionality and extending useful features to a diverse, majority voting based ensemble.

Customer churn prediction for telecommunication: Employing various various features selection techniques and tree based ensemble classifiers

Ensemble classifiers have received increasing attention for attaining the higher classification performance in recent times. In this paper, we present comparative performances of various tree based ensemble classifiers in collaboration with maximum relevancy and minimum redundancy (mRMR), Fishers ratio and F-score based features selection schemes for a challenging problem of churn prediction in telecommunication. The large sized telecommunication dataset has been the main hurdle in achieving the desired classification performance in the contemporary proposed churn prediction models. Though, tree based ensemble classifiers are considered suitable for larger datasets, but we have found rotation forest and rotboost as effective techniques compared to random forest, which employ boosting through features selection and increased diversity by incorporating linear feature extraction method such as Principal Component Analysis. In addition to the features selection performed by used ensembles, we have also incorporated mRMR, Fishers ratio and F-score techniques for features selection. mRMR returns a coherent and well discriminants feature set, compared to Fishers ratio and F-score, which significantly reduces the computations and helps classifier in attaining improved performance. The performance evaluation is conducted using area under curve, sensitivity and specificity where Rotboost, an ensemble of rotation forest and Adaboost in collaboration with mRMR has shown competitive results for churn prediction in telecommunication as compared to other ensemble methods.

An ensemble classification approach for automated diagnosis of Alzheimer's disease and mild cognitive impairment

Effective and robust detection of Alzheimer disease (AD) and mild cognitive impairment (MCI), which is a prodromal stage of AD, has attracted a lot of attention in the recent years. So far, multiple feature types such as hippocampal volume and shape, voxel based morphometry, and cortical thickness based features extracted from structural magnetic resonance imaging (MRI) have been proved to be effective for the diagnosis of MCI and AD. This work specifically deals with the features calculated from cortical surface. In particular, the techniques based on regional features of cortical surface poorly reveal the spatial variation in the cortical thickness, and the features defined on vertices are noise sensitive due to the large number of dimensions in MRI data. In this study, we present an ensemble classification framework developed on the average and vertex-wise features of cortical thickness. Furthermore, we also combine thickness based features with volume based features of cortex. The features were reduced using F-Score feature selection method prior to classification. The proposed technique has been tested on a dataset of 60 AD, 60 MCI, and 60 normal controls (NC) from Alzheimer Disease Neuro imaging Initiative (ADNI) dataset, and improvement in performance has been observed compared to previously reported studies in terms of sensitivity, specificity, accuracy and kappa statistics. The results show that the proposed feature selection coupled with ensemble classification significantly improves performance.

Earthquake prediction model using support vector regressor and hybrid neural networks

Earthquake prediction has been a challenging research area, where a future occurrence of the devastating catastrophe is predicted. In this work, sixty seismic features are computed through employing seismological concepts, such as Gutenberg-Richter law, seismic rate changes, foreshock frequency, seismic energy release, total recurrence time. Further, Maximum Relevance and Minimum Redundancy (mRMR) criteria is applied to extract the relevant features. A Support Vector Regressor (SVR) and Hybrid Neural Network (HNN) based classification system is built to obtain the earthquake predictions. HNN is a step wise combination of three different Neural Networks, supported by Enhanced Particle Swarm Optimization (EPSO), to offer weight optimization at each layer. The newly computed seismic features in combination with SVR-HNN prediction system is applied on Hindukush, Chile and Southern California regions. The obtained numerical results show improved prediction performance for all the considered regions, compared to previous prediction studies.