scholarly journals Deep learning framework for handling concept drift and class imbalanced complex decision-making on streaming data

2021 ◽  
Author(s):  
S. Priya ◽  
R. Annie Uthra
Keyword(s):  
Decision Making ◽  
Deep Learning ◽  
Concept Drift ◽  
Class Imbalance ◽  
Streaming Data ◽  
Superior Performance ◽  
Data Streaming ◽  
Minority Class ◽  
Concept Drift Detection

AbstractIn present times, data science become popular to support and improve decision-making process. Due to the accessibility of a wide application perspective of data streaming, class imbalance and concept drifting become crucial learning problems. The advent of deep learning (DL) models finds useful for the classification of concept drift in data streaming applications. This paper presents an effective class imbalance with concept drift detection (CIDD) using Adadelta optimizer-based deep neural networks (ADODNN), named CIDD-ADODNN model for the classification of highly imbalanced streaming data. The presented model involves four processes namely preprocessing, class imbalance handling, concept drift detection, and classification. The proposed model uses adaptive synthetic (ADASYN) technique for handling class imbalance data, which utilizes a weighted distribution for diverse minority class examples based on the level of difficulty in learning. Next, a drift detection technique called adaptive sliding window (ADWIN) is employed to detect the existence of the concept drift. Besides, ADODNN model is utilized for the classification processes. For increasing the classifier performance of the DNN model, ADO-based hyperparameter tuning process takes place to determine the optimal parameters of the DNN model. The performance of the presented model is evaluated using three streaming datasets namely intrusion detection (NSL KDDCup) dataset, Spam dataset, and Chess dataset. A detailed comparative results analysis takes place and the simulation results verified the superior performance of the presented model by obtaining a maximum accuracy of 0.9592, 0.9320, and 0.7646 on the applied KDDCup, Spam, and Chess dataset, respectively.

Concept Drift Detection on Streaming Data with Dynamic Outlier Aggregation

2021 ◽  
pp. 206-217
Author(s):  
Ludwig Zellner ◽  
Florian Richter ◽  
Janina Sontheim ◽  
Andrea Maldonado ◽  
Thomas Seidl
Keyword(s):  
Concept Drift ◽  
Streaming Data ◽  
Concept Drift Detection

scholarly journals Data-driven decision support under concept drift in streamed big data

2019 ◽  
Vol 6 (1) ◽  
pp. 157-163 ◽  
Author(s):  
Jie Lu ◽  
Anjin Liu ◽  
Yiliao Song ◽  
Guangquan Zhang
Keyword(s):  
Decision Making ◽  
Big Data ◽  
Real Time ◽  
Concept Drift ◽  
High Volume ◽  
Streaming Data ◽  
Data Driven ◽  
Research Directions ◽  
Decision Outcomes ◽  
Past Data

Abstract Data-driven decision-making ($$\mathrm {D^3}$$D3M) is often confronted by the problem of uncertainty or unknown dynamics in streaming data. To provide real-time accurate decision solutions, the systems have to promptly address changes in data distribution in streaming data—a phenomenon known as concept drift. Past data patterns may not be relevant to new data when a data stream experiences significant drift, thus to continue using models based on past data will lead to poor prediction and poor decision outcomes. This position paper discusses the basic framework and prevailing techniques in streaming type big data and concept drift for $$\mathrm {D^3}$$D3M. The study first establishes a technical framework for real-time $$\mathrm {D^3}$$D3M under concept drift and details the characteristics of high-volume streaming data. The main methodologies and approaches for detecting concept drift and supporting $$\mathrm {D^3}$$D3M are highlighted and presented. Lastly, further research directions, related methods and procedures for using streaming data to support decision-making in concept drift environments are identified. We hope the observations in this paper could support researchers and professionals to better understand the fundamentals and research directions of $$\mathrm {D^3}$$D3M in streamed big data environments.

scholarly journals Classification of Hyperspectral Image Based on Double-Branch Dual-Attention Mechanism Network

2020 ◽  
Vol 12 (3) ◽  
pp. 582 ◽  
Author(s):  
Rui Li ◽  
Shunyi Zheng ◽  
Chenxi Duan ◽  
Yang Yang ◽  
Xiqi Wang
Keyword(s):  
Deep Learning ◽  
Hyperspectral Image ◽  
State Of The Art ◽  
Attention Mechanism ◽  
Superior Performance ◽  
Feature Maps ◽  
Spatial Features ◽  
Training Samples ◽  
Series Of Experiments

In recent years, researchers have paid increasing attention on hyperspectral image (HSI) classification using deep learning methods. To improve the accuracy and reduce the training samples, we propose a double-branch dual-attention mechanism network (DBDA) for HSI classification in this paper. Two branches are designed in DBDA to capture plenty of spectral and spatial features contained in HSI. Furthermore, a channel attention block and a spatial attention block are applied to these two branches respectively, which enables DBDA to refine and optimize the extracted feature maps. A series of experiments on four hyperspectral datasets show that the proposed framework has superior performance to the state-of-the-art algorithm, especially when the training samples are signally lacking.

scholarly journals A Dynamic Ensemble Framework for Mining Textual Streams with Class Imbalance

2014 ◽  
Vol 2014 ◽  
pp. 1-11 ◽  
Author(s):  
Ge Song ◽  
Yunming Ye
Keyword(s):  
Large Scale ◽  
State Of The Art ◽  
Concept Drift ◽  
Real Life ◽  
Class Imbalance ◽  
High Dimensional ◽  
Adaptive Selection ◽  
Stream Classification ◽  
Rare Class

Textual stream classification has become a realistic and challenging issue since large-scale, high-dimensional, and non-stationary streams with class imbalance have been widely used in various real-life applications. According to the characters of textual streams, it is technically difficult to deal with the classification of textual stream, especially in imbalanced environment. In this paper, we propose a new ensemble framework, clustering forest, for learning from the textual imbalanced stream with concept drift (CFIM). The CFIM is based on ensemble learning by integrating a set of clustering trees (CTs). An adaptive selection method, which flexibly chooses the useful CTs by the property of the stream, is presented in CFIM. In particular, to deal with the problem of class imbalance, we collect and reuse both rare-class instances and misclassified instances from the historical chunks. Compared to most existing approaches, it is worth pointing out that our approach assumes that both majority class and rareclass may suffer from concept drift. Thus the distribution of resampled instances is similar to the current concept. The effectiveness of CFIM is examined in five real-world textual streams under an imbalanced nonstationary environment. Experimental results demonstrate that CFIM achieves better performance than four state-of-the-art ensemble models.

scholarly journals K-means-SMOTE for handling class imbalance in the classification of diabetes with C4.5, SVM, and naive Bayes

2020 ◽  
Vol 8 (2) ◽  
pp. 89-93 ◽  
Author(s):  
Hairani Hairani ◽  
Khurniawan Eko Saputro ◽  
Sofiansyah Fadli
Keyword(s):  
Naive Bayes ◽  
Class Imbalance ◽  
Naïve Bayes ◽  
Data Sampling ◽  
Minority Class ◽  
Class A ◽  
Positive Class ◽  
Negative Class ◽  
Imbalanced Class

The occurrence of imbalanced class in a dataset causes the classification results to tend to the class with the largest amount of data (majority class). A sampling method is needed to balance the minority class (positive class) so that the class distribution becomes balanced and leading to better classification results. This study was conducted to overcome imbalanced class problems on the Indian Pima diabetes illness dataset using k-means-SMOTE. The dataset has 268 instances of the positive class (minority class) and 500 instances of the negative class (majority class). The classification was done by comparing C4.5, SVM, and naïve Bayes while implementing k-means-SMOTE in data sampling. Using k-means-SMOTE, the SVM classification method has the highest accuracy and sensitivity of 82 % and 77 % respectively, while the naive Bayes method produces the highest specificity of 89 %.

scholarly journals Label-free imaging and classification of live P. falciparum

2020 ◽  
Author(s):  
Paul M. Lebel ◽  
Rebekah L. Dial ◽  
Venkata N. P. Vemuri ◽  
Valentina Garcia ◽  
Joseph DeRisi ◽  
...  
Keyword(s):  
Deep Learning ◽  
Limit Of Detection ◽  
Life Stage ◽  
Superior Performance ◽  
Label Free ◽  
Near Ultraviolet ◽  
Smear Preparation ◽  
Blood Smears ◽  
Microscopic Inspection

AbstractManual microscopic inspection of fixed and stained blood smears has remained the gold standard for Plasmodium parasitemia analysis for over a century. Unfortunately, smear preparation consumes time and reagents, while manual microscopy is skill-dependent and labor-intensive. Here, we demonstrate that label-free microscopy combined with deep learning enables both life stage classification and accurate parasitemia quantification. Using a custom-built microscope, we find that deep-ultraviolet light enhances image contrast and resolution, achieving four-category classification of Plasmodium falciparum blood stages at an overall accuracy greater than 99%. To increase accessibility, we extended our method to a commercial brightfield microscope using near-ultraviolet and visible light. Both systems were tested extrinsically by parasitemia titration, revealing superior performance over manually-scored Giemsa-stained smears, and a limit of detection below 0.1%. Our results suggest that label-free microscopy combined with deep learning could eliminate the need for conventional blood smear analysis.

scholarly journals Continuous Training and Deployment of Deep Learning Models

2021 ◽  
Author(s):  
Ioannis Prapas ◽  
Behrouz Derakhshan ◽  
Alireza Rezaei Mahdiraji ◽  
Volker Markl
Keyword(s):  
Deep Learning ◽  
Historical Data ◽  
State Of The Art ◽  
Streaming Data ◽  
Superior Performance ◽  
Learning Models ◽  
Model Quality ◽  
Continuous Training ◽  
Training Time ◽  
Machine Learning Methods

AbstractDeep Learning (DL) has consistently surpassed other Machine Learning methods and achieved state-of-the-art performance in multiple cases. Several modern applications like financial and recommender systems require models that are constantly updated with fresh data. The prominent approach for keeping a DL model fresh is to trigger full retraining from scratch when enough new data are available. However, retraining large and complex DL models is time-consuming and compute-intensive. This makes full retraining costly, wasteful, and slow. In this paper, we present an approach to continuously train and deploy DL models. First, we enable continuous training through proactive training that combines samples of historical data with new streaming data. Second, we enable continuous deployment through gradient sparsification that allows us to send a small percentage of the model updates per training iteration. Our experimental results with LeNet5 on MNIST and modern DL models on CIFAR-10 show that proactive training keeps models fresh with comparable—if not superior—performance to full retraining at a fraction of the time. Combined with gradient sparsification, sparse proactive training enables very fast updates of a deployed model with arbitrarily large sparsity, reducing communication per iteration up to four orders of magnitude, with minimal—if any—losses in model quality. Sparse training, however, comes at a price; it incurs overhead on the training that depends on the size of the model and increases the training time by factors ranging from 1.25 to 3 in our experiments. Arguably, a small price to pay for successfully enabling the continuous training and deployment of large DL models.

scholarly journals Data stream mining: methods and challenges for handling concept drift

2019 ◽  
Vol 1 (11) ◽  
Author(s):  
Scott Wares ◽  
John Isaacs ◽  
Eyad Elyan
Keyword(s):  
Data Stream ◽  
Concept Drift ◽  
Relevant Literature ◽  
Streaming Data ◽  
Future Research ◽  
Stream Mining ◽  
Detection Algorithms ◽  
The Past ◽  
Concept Drift Detection ◽  
The Impact

Abstract Mining and analysing streaming data is crucial for many applications, and this area of research has gained extensive attention over the past decade. However, there are several inherent problems that continue to challenge the hardware and the state-of-the art algorithmic solutions. Examples of such problems include the unbound size, varying speed and unknown data characteristics of arriving instances from a data stream. The aim of this research is to portray key challenges faced by algorithmic solutions for stream mining, particularly focusing on the prevalent issue of concept drift. A comprehensive discussion of concept drift and its inherent data challenges in the context of stream mining is presented, as is a critical, in-depth review of relevant literature. Current issues with the evaluative procedure for concept drift detectors is also explored, highlighting problems such as a lack of established base datasets and the impact of temporal dependence on concept drift detection. By exposing gaps in the current literature, this study suggests recommendations for future research which should aid in the progression of stream mining and concept drift detection algorithms.

Sign in / Sign up

Export Citation Format

Share Document