A Comparative Analysis of Convergence Rate for Imbalanced Datasets of Active Learning Models

Ecological research like wildlife censuses increasingly relies on data on the scale of Terabytes. For example, modern camera trap datasets contain millions of images that require prohibitive amounts of manual labour to be annotated with species, bounding boxes, and the like. Machine learning, especially deep learning [3], could greatly accelerate this task through automated predictions, but involves expansive coding and expert knowledge.In this abstract we present AIDE, the Annotation Interface for Data-driven Ecology [2]. In a first instance, AIDE is a web-based annotation suite for image labelling with support for concurrent access and scalability, up to the cloud. In a second instance, it tightly integrates deep learning models into the annotation process through active learning [7], where models learn from user-provided labels and in turn select the most relevant images for review from the large pool of unlabelled ones (Fig. 1). The result is a system where users only need to label what is required, which saves time and decreases errors due to fatigue.<img src="https://contentmanager.copernicus.org/fileStorageProxy.php?f=gnp.0402be60f60062057601161/sdaolpUECMynit/12UGE&app=m&a=0&c=131251398e575ac9974634bd0861fadc&ct=x&pn=gnp.elif&d=1" alt="">Fig. 1: AIDE offers concurrent web image labelling support and uses annotations and deep learning models in an active learning loop.AIDE includes a comprehensive set of built-in models, such as ResNet [1] for image classification, Faster R-CNN [5] and RetinaNet [4] for object detection, and U-Net [6] for semantic segmentation. All models can be customised and used without having to write a single line of code. Furthermore, AIDE accepts any third-party model with minimal implementation requirements. To complete the package, AIDE offers both user annotation and model prediction evaluation, access control, customisable model training, and more, all through the web browser.AIDE is fully open source and available under https://github.com/microsoft/aerial_wildlife_detection.&#160;References

Download Full-text

Comparative Analysis of Machine Learning Models for the Prediction of Pedestrian Crash Severity: Focused on Balancing Pedestrian Crash Dataset

Journal of Korean Society for Geospatial Information System ◽

10.7319/kogsis.2021.29.2.003 ◽

2021 ◽

Vol 29 (2) ◽

pp. 3-15

Author(s):

Hojun Lee ◽

Sugie Lee

Keyword(s):

Machine Learning ◽

Comparative Analysis ◽

Crash Severity ◽

Learning Models ◽

Machine Learning Models

Download Full-text

Motivational Active Learning in Blended and Virtual Learning Scenarios

Online Course Management ◽

10.4018/978-1-5225-5472-1.ch047 ◽

2018 ◽

pp. 935-957

Author(s):

Johanna Pirker ◽

Maria Riffnaller-Schiefer ◽

Lisa Maria Tomes ◽

Christian Gütl

Keyword(s):

Active Learning ◽

Learning Strategies ◽

Relevant Literature ◽

Virtual Learning ◽

Learning Approaches ◽

Learning Models ◽

Engagement Strategies ◽

Active And Collaborative Learning ◽

Learning Success ◽

Learning Scenarios

The way people learn has changed over the last years. New pedagogical theories show that engaging and active learning approaches are particularly successful in improving conceptual understanding and enhancing the students' learning success and motivation. The Motivational Active Learning approach combines engagement strategies based on active and collaborative learning models with gamification. While many active learning models rely on in-class setups and active and personal interactions between students and between instructors, MAL was designed to integrate active learning in different settings. Our research project focuses on enhanced learning strategies with MAL in different computer-supported scenarios. This chapter outlines the potential of the pedagogical model MAL (Motivational Active Learning) in the context of blended and virtual learning scenarios; it also summarizes relevant literature and discusses implications and future work.

Download Full-text

Improving Logging Prediction on Imbalanced Datasets

International Journal of Open Source Software and Processes ◽

10.4018/ijossp.2016040103 ◽

2016 ◽

Vol 7 (2) ◽

pp. 43-71 ◽

Cited By ~ 3

Author(s):

Sangeeta Lal ◽

Neetu Sardana ◽

Ashish Sureka

Keyword(s):

Machine Learning ◽

Open Source ◽

Class Imbalance ◽

Learning Model ◽

Learning Models ◽

Class Imbalance Problem ◽

Imbalanced Datasets ◽

Imbalance Problem ◽

Machine Learning Model ◽

Machine Learning Models

Logging is an important yet tough decision for OSS developers. Machine-learning models are useful in improving several steps of OSS development, including logging. Several recent studies propose machine-learning models to predict logged code construct. The prediction performances of these models are limited due to the class-imbalance problem since the number of logged code constructs is small as compared to non-logged code constructs. No previous study analyzes the class-imbalance problem for logged code construct prediction. The authors first analyze the performances of J48, RF, and SVM classifiers for catch-blocks and if-blocks logged code constructs prediction on imbalanced datasets. Second, the authors propose LogIm, an ensemble and threshold-based machine-learning model. Third, the authors evaluate the performance of LogIm on three open-source projects. On average, LogIm model improves the performance of baseline classifiers, J48, RF, and SVM, by 7.38%, 9.24%, and 4.6% for catch-blocks, and 12.11%, 14.95%, and 19.13% for if-blocks logging prediction.

Download Full-text