Filters








271 Hits in 2.7 sec

Deep Learning in Customer Churn Prediction: Unsupervised Feature Learning on Abstract Company Independent Feature Vectors [article]

Philip Spanoudes, Thomson Nguyen
2017 arXiv   pre-print
The paper describes in depth the application of Deep Learning in the problem of churn prediction.  ...  secondary features in an unsupervised manner), the complete pipeline can be applied to any subscription based company with extremely good churn predictive performance.  ...  The authors would like to acknowledge Elliot Block, Andrew Berls, Dan Evans, and the rest of the Framed Data team for their engineering and data infrastructure assistance and technical guidance.  ... 
arXiv:1703.03869v1 fatcat:hrpzo3pt2balpo3zm5it3dmmi4

Experimental Parameter Tuning of Artificial Neural Network in Customer Churn Prediction

Martin Fridrich
2017 Trendy Ekonomiky a Managementu  
Scientific aim: To present and execute experimental design for performance evaluation and hyperparameter optimization of classification models, which are used for customer churn prediction.  ...  Conclusions: Results imply that placing hyperparameter optimization to ANN classification model leads to improved customer churn prediction ability.  ...  Popular techniques used for customer churn prediction are logistic regression, decision tree, fuzzy logic, Bayesian classifier, SVM, and neural networks (Ngai et al., 2009; Kumar, Garg, 2013) , the paper  ... 
doi:10.13164/trends.2017.28.9 fatcat:3lzpxwvi4zf2vo2kh3g55xy6we

Customer Lifetime Value Prediction Using Embeddings

Benjamin Paul Chamberlain, Ângelo Cardoso, C.H. Bryan Liu, Roberto Pagliari, Marc Peter Deisenroth
2017 Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '17  
The state of the art in this domain uses large numbers of handcrafted features and ensemble regressors to forecast value, predict churn and evaluate customer loyalty.  ...  We describe the Customer LifeTime Value (CLTV) prediction system deployed at ASOS.com, a global online fashion retailer.  ...  We experiment with (1) deep feed-forward neural networks and (2) hybrid models combining logistic regression and a deep feedforward neural network similar to that used in [5] . e deep feedforward neural  ... 
doi:10.1145/3097983.3098123 dblp:conf/kdd/ChamberlainCLPD17 fatcat:vyosgrgxlvbhxkzg7ib26vikaa

I Know You'll Be Back: Interpretable New User Clustering and Churn Prediction on a Mobile Social Application [article]

Carl Yang, Xiaolin Shi, Jie Luo, Jiawei Han
2019 arXiv   pre-print
Then we design a novel deep learning pipeline based on LSTM and attention to accurately predict user churn with very limited initial behavior data, by leveraging the correlations among users' multi-dimensional  ...  The whole framework is deployed as a data analysis pipeline, delivering real-time data analysis and prediction results to multiple relevant teams for business intelligence uses.  ...  We deployed ClusChurn in Snap Inc. to deliver real-time data analysis and prediction results to benefit multiple productions including user modeling, growth, retention, and so on.  ... 
arXiv:1910.01447v1 fatcat:mfea4xsmgffhxnc6diyjizxe2u

Game Data Mining Competition on Churn Prediction and Survival Analysis using Commercial Game Log Data

Eunjo Lee, Yoonjae Jang, Du-Mim Yoon, JiHoon Jeon, Sung-il Yang, SangKwang Lee, Dae-Wook Kim, Pei Pei Chen, Anna Guitart, Paul Bertens, Africa Perianez, Fabian Hadiji (+5 others)
2019 IEEE Transactions on Games  
The main aim of the competition was to predict whether a player would churn and when the player would churn during two periods between which the business model was changed to a free-to-play model from  ...  The results of the competition revealed that highly ranked competitors used deep learning, tree boosting, and linear regression.  ...  Parameters were adjusted through cross-validation. 1) Binary Churn Prediction (Track 1) 1-1) LSTM A model combining an LSTM network with a deep neural network (DNN) was used for test set 1 ( Figure  ... 
doi:10.1109/tg.2018.2888863 fatcat:ifaukda6arhsjiz65jaasfjlly

Adaptive XGBOOST Hyper Tuned Meta Classifier for Prediction of Churn Customers

B. Srikanth, Swarajya Lakshmi V. Papineni, Gutta Sridevi, D. N. V. S. L. S. Indira, K. S. R. Radhika, Khasim Syed
2022 Intelligent Automation and Soft Computing  
Churn prediction is the number of customers who wants to terminate their services in the banking sector.  ...  The model considers twelve attributes like credit score, geography, gender, age, etc, to predict customer churn. The project consists of five modules as follows.  ...  This model uses Multi-Layer Perceptron in Artificial Neural Network (ANN) [9] for customer churn prediction and resulted in the best accuracy compared to the traditional machine learning methods with  ... 
doi:10.32604/iasc.2022.022423 fatcat:rtpffzi4qzfw7cjwwtep3ljmhm

Game Data Mining Competition on Churn Prediction and Survival Analysis using Commercial Game Log Data

Kyung-Joong Kim, DuMim Yoon, JiHoon Jeon, Seong-Il Yang, Sang-Kwang Lee, EunJo Lee, Yoonjae Jang, Dae-Wook Kim, Pei Pei Chen, Anna Guitart, Paul Bertens, África Periáñez (+5 others)
2018 Zenodo  
The main aim of the competition was to predict whether a player would churn and when the player would churn during two periods between which the business model was changed to a free-to-play model from  ...  The results of the competition revealed that highly ranked competitors used deep learning, tree boosting, and linear regression.  ...  Parameters were adjusted through cross-validation. 1) Binary Churn Prediction (Track 1) 1-1) LSTM A model combining an LSTM network with a deep neural network (DNN) was used for test set 1 ( Figure  ... 
doi:10.5281/zenodo.1696795 fatcat:rejyi4n7f5ht3l7cmjoiax4fgu

Analysis of WTTE-RNN Variants that Improve Performance

Rory Cawley, Dr.John Burns
2019 Machine Learning and Applications An International Journal  
A prediction model called the WTTE-RNN was shown to be effective for predicting the time to event for topics such as machine failure.  ...  These assets share a common trait in that at some stage they will fail or, in the case of customers, they will churn.  ...  The focus of the research explores using an RNN for predicting churn based on customer lifetime value (CLV) time series. C. PREDICTING CUSTOMER CHURN USING RECURRENT NEURAL NET-WORKS D.  ... 
doi:10.5121/mlaij.2019.6103 fatcat:5t6qapvx2jehpopcvrbaa2ea44

Towards Searching Efficient and Accurate Neural Network Architectures in Binary Classification Problems [article]

Yigit Alparslan, Ethan Jacob Moyer, Isamu Mclean Isozaki, Daniel Schwartz, Adam Dunlop, Shesh Dave, Edward Kim
2021 arXiv   pre-print
In recent years, deep neural networks have had great success in machine learning and pattern recognition.  ...  Architecture size for a neural network contributes significantly to the success of any neural network.  ...  ACKNOWLEDGMENT We would like to acknowledge Drexel Society of Artificial Intelligence for its contributions and support for this research.  ... 
arXiv:2101.06511v1 fatcat:y5xjh6aktbfspjn7rli4rfx2iq

Randomness In Neural Network Training: Characterizing The Impact of Tooling [article]

Donglin Zhuang, Xingyao Zhang, Shuaiwen Leon Song, Sara Hooker
2021 arXiv   pre-print
In this work, we address a less well understood and studied question: how does our choice of tooling introduce randomness to deep neural network training.  ...  Our results suggest that deterministic tooling is critical for AI safety.  ...  Churn (churn)-Predictive churn is a measure of predictive divergence between two models.  ... 
arXiv:2106.11872v1 fatcat:7shulcpu3bb3tikwl3d3naweku

Large scale distributed neural network training through online distillation [article]

Rohan Anil, Gabriel Pereyra, Alexandre Passos, Robert Ormandi, George E. Dahl, Geoffrey E. Hinton
2020 arXiv   pre-print
However, due to increased test-time cost (for ensembles) and increased complexity of the training pipeline (for distillation), these techniques are challenging to use in industrial settings.  ...  Two neural networks trained on disjoint subsets of the data can share knowledge by encouraging each model to agree with the predictions the other model would have made.  ...  ACKNOWLEDGMENTS We would like to thank Avital Oliver for feedback on a draft and Oriol Vinyals for many helpful discussions.  ... 
arXiv:1804.03235v2 fatcat:ftkf2wofpjhqxaeilzksiwpfsq

RICON: A ML framework for real-time and proactive intervention to prevent customer churn [article]

Arnab Chakraborty, Vikas Raturi, Shrutendra Harsola
2022 arXiv   pre-print
We consider the problem of churn prediction in real-time.  ...  In addition to churn propensity prediction, RICON provides insights based on product usage intelligence.  ...  for churn detection in telecom industries [1, 9] ; difficulty-aware churn prediction framework [15] and survival ensemble-based multi-dimensional churn prediction in online gaming [2] ; Deep-CNN and  ... 
arXiv:2203.16155v1 fatcat:6en2ry4e3jacdbt3adwiylwbfq

Visual Reasoning of Feature Attribution with Deep Recurrent Neural Networks [article]

Chuan Wang, Takeshi Onishi, Keiichi Nemoto, Kwan-Liu Ma
2019 arXiv   pre-print
Deep Recurrent Neural Network (RNN) has gained popularity in many sequence classification tasks.  ...  Beyond predicting a correct class for each data instance, data scientists also want to understand what differentiating factors in the data have contributed to the classification during the learning process  ...  RELATED WORK Visualization for Deep Neural Networks (DNN) Many visualization techniques have been developed to facilitate the DNN model building process, covering domains such as image understanding [  ... 
arXiv:1901.05574v1 fatcat:5ikq3zzflzdqffv2wmsx3qikru

A Blended Deep Learning Approach for Predicting User Intended Actions [article]

Fei Tan, Zhi Wei, Jun He, Xiang Wu, Bo Peng, Haoran Liu, Zhenyu Yan
2018 arXiv   pre-print
To overcome these limitations, we propose a novel end-to-end learning scheme to keep track of the evolution of attrition patterns for the predictive modeling.  ...  Conventional attrition predictive modeling strategies suffer a few inherent drawbacks.  ...  ACKNOWLEDGMENT We thank Sagar Patil for proofreading the manuscript.  ... 
arXiv:1810.04824v1 fatcat:6okce4rlrrd6hn4bpot7zfvmbi

Assessment of an Identification Strategy for the Prediction of the Dynamics of Two-phase Flows

A. Fichera, A. Pagano
2015 Energy Procedia  
Such a model has been implemented by means of Multilayer Perceptron artificial neural networks, trained using input-output data detected during experiments expressing different flow patterns.  ...  The proposed strategy consists in the assessment and optimisation of a generalised NARMAX model for the input-output identification of the dynamics of the experimental time series of the void fraction,  ...  to be required for the attractors of more complex flow patterns, such as those of churn and annular flows.  ... 
doi:10.1016/j.egypro.2015.12.039 fatcat:xi2yk7e3zrhjxhrlfwog6aonju
« Previous Showing results 1 — 15 out of 271 results