Learning from Unbalanced Stream Data in Non-Stationary Environments Using Logistic Regression Model

Author(s):  
Pallavi Digambarrao Kulkarni ◽  
Roshani Ade

There are several deep learning approaches that can be applied for analyzing situations in real world problems and inventing their solution in a scientific technique. Supervised data mining methods that predicts instance values, using previously obtained results from already collected data are pretty popular due to their intelligence in machine learning area. Stream data is continuous form of data which can be handled by using incremental learning approach. Stream data learning may face several challenges in real world like concept drift or class imbalance. Concept drift occurs in non-stationary environment where data distribution generation function is dynamic in nature and has no fixed formula to predict the future data distribution nature. Neural network techniques are intelligent enough to improve performance of algorithmic systems that work in such problem domains. This chapter briefly describes how MLP technique is integrated in system so that the system becomes a complete framework for handling unbalanced data with concept drift in the incremental learning strategies.

Author(s):  
Pallavi Digambarrao Kulkarni ◽  
Roshani Ade

There are several deep learning approaches that can be applied for analyzing situations in real world problems and inventing their solution in a scientific technique. Supervised data mining methods that predicts instance values, using previously obtained results from already collected data are pretty popular due to their intelligence in machine learning area. Stream data is continuous form of data which can be handled by using incremental learning approach. Stream data learning may face several challenges in real world like concept drift or class imbalance. Concept drift occurs in non-stationary environment where data distribution generation function is dynamic in nature and has no fixed formula to predict the future data distribution nature. Neural network techniques are intelligent enough to improve performance of algorithmic systems that work in such problem domains. This chapter briefly describes how MLP technique is integrated in system so that the system becomes a complete framework for handling unbalanced data with concept drift in the incremental learning strategies.


2021 ◽  
Vol 9 (2) ◽  
pp. 36-52
Author(s):  
Mashaal A. Alfhaid ◽  
Manal Abdullah

As the number of generated data increases every day, this has brought the importance of data mining and knowledge extraction. In traditional data mining, offline status can be used for knowledge extraction. Nevertheless, dealing with stream data mining is different due to continuously arriving data that can be processed at a single scan besides the appearance of concept drift. As the pre-processing stage is critical in knowledge extraction, imbalanced stream data gain significant popularity in the last few years among researchers. Many real-world applications suffer from class imbalance including medical, business, fraud detection and etc. Learning from the supervised model includes classes whether it is binary- or multi-classes. These classes are often imbalance where it is divided into the majority (negative) class and minority (positive) class, which can cause a bias toward the majority class that leads to skew in predictive performance models. Handles imbalance streaming data is mandatory for more accurate and reliable learning models. In this paper, we will present an overview of data stream mining and its tools. Besides, summarize the problem of class imbalance and its different approaches. In addition, researchers will present the popular evaluation metrics and challenges prone from imbalanced streaming data.


Information ◽  
2019 ◽  
Vol 10 (5) ◽  
pp. 158 ◽  
Author(s):  
Yange Sun ◽  
Han Shao ◽  
Shasha Wang

Most existing multi-label data streams classification methods focus on extending single-label streams classification approaches to multi-label cases, without considering the special characteristics of multi-label stream data, such as label dependency, concept drift, and recurrent concepts. Motivated by these challenges, we devise an efficient ensemble paradigm for multi-label data streams classification. The algorithm deploys a novel change detection based on Jensen–Shannon divergence to identify different kinds of concept drift in data streams. Moreover, our method tries to consider label dependency by pruning away infrequent label combinations to enhance classification performance. Empirical results on both synthetic and real-world datasets have demonstrated its effectiveness.


Cubic Journal ◽  
2021 ◽  
pp. 44-53
Author(s):  
Gladys Lam Wai Ling

Technological developments have brought profound challenges to design education. To understand how design educators adapt to new technological directions, this article examines student feedback from advertising design courses that apply blended learning approaches. This study identified three blended learning strategies conducive to meaningful learning: timely and meaningful feedback; engagement with real world tasks; and support from expert tutors. This article also discusses potential resistance and challenges in implementing instruction in blended technological environments.


Author(s):  
Yang Lu ◽  
Yiu-ming Cheung ◽  
Yuan Yan Tang

Concept drifts occurring in data streams will jeopardize the accuracy and stability of the online learning process. If the data stream is imbalanced, it will be even more challenging to detect and cure the concept drift. In the literature, these two problems have been intensively addressed separately, but have yet to be well studied when they occur together. In this paper, we propose a chunk-based incremental learning method called Dynamic Weighted Majority for Imbalance Learning (DWMIL) to deal with the data streams with concept drift and class imbalance problem. DWMIL utilizes an ensemble framework by dynamically weighting the base classifiers according to their performance on the current data chunk. Compared with the existing methods, its merits are four-fold: (1) it can keep stable for non-drifted streams and quickly adapt to the new concept; (2) it is totally incremental, i.e. no previous data needs to be stored; (3) it keeps a limited number of classifiers to ensure high efficiency; and (4) it is simple and needs only one thresholding parameter. Experiments on both synthetic and real data sets with concept drift show that DWMIL performs better than the state-of-the-art competitors, with less computational cost.


Author(s):  
Célia Quintas ◽  
Ana Luísa Teixeira ◽  
Isabel Fernandes Silva ◽  
Jane Rodrigues Duarte

Knowledge management and learning are buzzwords in today’s society, both in terms of company competitiveness as well as in terms of education. Human resources are thus a priority for individuals and companies. The concept of knowledge management and of learning organizations has been object of increased interest by managers and scholars. The increased focus on these issues brings forth the individual as a crucial element in this process; individuals become key elements in competitiveness (Nonaka & Takeuchi: 1995) and protagonists of their own learning process (Senge: 1992).Additionally, the learning methodologies and strategies have also changed in the past decades, so that currently much is offered by means of b-learning and e-learning courses that, on the one hand, allow students to opt for several learning strategies, and on the other hand, require them to actively participate in their learning path. In fact, the evolution of ICT in studies and the growing experience of both teachers and students have gradually adapted to new methodologies. However, while materials and subject matter have been made easier and more accessible to students who do not attend classroom sessions, an underlying problem has always been present: bridging the physical distance among all the stakeholders involved in the learning process and all the difficulties that may emerge from this.Since its first edition in 2001, this Post-Graduation Program, now in its 12th edition, has undergone several changes, from its study plan to learning regime. As a means of responding to the demands of today’s market and in particular new learning styles, new possibilities have been made for attending the course which range from classroom, to blending and e-learning formats. As a means of fostering group spirit, synchronous and asynchronous participation of all students several changes were introduced this academic year. Besides the use of the Moodle platform, a Virtual Learning Environment (VLE) wiziq has been introduced.In 2013-14, the program includes students from Portugal (including the Azores), Mexico and Nigeria. Moreover, this Post-Graduation Program allows students to opt for f2f, b-learning and e-learning regimes, i.e., within the same group, some students attend classes by means of a VLE, others attend some classes f2f and others using the VLE and others attend f2f classes regularly, though they also have access to the VLE. A program that combines three learning approaches/methodologies/strategies allows the possibility of assessing possible differences in terms of efficiency of these three learning methodologies, considering that these imply a change in expectations, attitude and cognitive process.Our paper focuses on a study carried out in a Post-Graduation Program at a Portuguese university, on perceived satisfaction regarding the use of ICT tools in the program, a theme which has already been object of study at UAL in recent years, both in terms of assessing and monitoring learning progress, of learner attitude toward their learning paths (Fernandes Silva & Rodrigues Duarte. 2011a & b) and the tools and methodologies made available to them and of perceived satisfaction (Fernandes Silva & Quintas: 2013).This paper corresponds to a 1st stage of a broader study that will involve all students in the referred program in 2013-14 as well as all the lecturers. Initially, a qualitative analysis is carried out based on semi-structured interviews; at a 2nd stage, we aim to create a questionnaire to be applied to a wider population.


Author(s):  
Erna Pebriana ◽  
Bela Mustika Sari ◽  
Yasa Abdurrahman

This writing aims to make students more active and disciplined in the learning process and can also increase creativity and learning outcomes. The low mathematics learning outcomes are not only due to difficult mathematics, but are caused by several factors which include students themselves, teachers, learning approaches, and learning environments that are interconnected with each other. To improve the ability and results of learning it is necessary to make modifications to the task learning strategy and force. Quantum learning is a tip, a guide, a strategy and an entire learning process that can sharpen understanding and memory, and make learning a pleasant and useful process. Task and Forced Learning Strategies are strategies that focus on giving assignments and a little coercion so that students complete their tasks on time so that the learning process can run effectively. Therefore, the writer modifies the model of quantum learning with task and forced learning strategies, the results of this modification show that learning with quantum learning models with forced and task strategies can improve the learning process so that students become more disciplined in doing tasks, can motivate student learning, and can improve student learning outcomes.


2020 ◽  
Vol 2 (1) ◽  
pp. 45-54
Author(s):  
Hikmah ◽  
Ance Jusmaya

Being a housewife is a multi-tasking  tasks and it is not an easy thing. In this case, a housewife has many roles such as should be a mother , a counselor for her daughter  as well as taking care of everything. Besides, the mother is also a teacher. As we know that,  the  first  teacher of a child is a mother. Then,  the mother is also a financial manager and general administration  at home. Many problems have been encountered, so a housewife  tasks are  very hard, in this case they have to  harmonize and regulate the amount of income and increase in some basic needs and daily needs. Except the problems that regarding  with financial management, the problem  face also relates with the lack of knowledge of housewives in English.  As a housewife needs an ability of English skill  to help their children  in studying later on.  Those phenomenon  happens in  families who live in Griya Batu Aji stage 1.The solution offered housewife  that a family financial management is very important for financial survival of a family. As a financial manager at home, a housewife must be able to manage expenditure and income posts. Besides, for teaching English,  parents should implement a fun learning environment and learning strategies that can motivate children to learn English. A learning environment that suits the real-world context is needed so that parents can apply it to everyday learning activities with children.


2021 ◽  
Vol 37 (1) ◽  
pp. 635-656
Author(s):  
Farzana Anowar ◽  
Samira Sadaoui

Sign in / Sign up

Export Citation Format

Share Document