Applying active learning strategy to classify large scale data with imbalanced classes

Phairod, Tuntiwachiratrakun; Vateekul, Peerapon

doi:10.1109/iccais.2016.7822443

Cited by 4 publications

(1 citation statement)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We hope these model performance gains, coupled with an iterative process of AL, will complement existing visual interpretation methods, thus accelerating tracking land-use following deforestation at a global scale on a wall-to-wall and more frequently. Similar improvements using AL has been reported in (Bengar et al, 2021;Kim et al, 2015;Ma et al, 2011;Tuntiwachiratrakun and Vateekul, 2017;Wang et al, 2017;Wu et al, 2014).…”

Section: Discussionsupporting

confidence: 81%

Assessing land use following tropical deforestation : Combining remote sensing and deep learning

Masolele

View full text Add to dashboard Cite

Forests are home to a vast variety species of plants and animals. They provide these species with vital ecosystem services such as fresh air, food, nutrients, and shade, essential for survival. They also help moderate the flow of freshwater and influence precipitation patterns on which humans depend for agriculture. Forests are also fundamental for removing carbon from the atmosphere through photosynthesis, and clearing them can also result in the emission of large amounts of carbon into the atmosphere. Thus, the protection and management of these forests are critical to maintaining a continual flow of these ecosystem services and supporting life on Earth.However, over the centuries, forests have been facing threats from human activities. About 50% of the Earth's land was covered with forests a thousand decades ago, half of which have since been cleared. Today, less than 33% of the remaining forests are still intact, mostly due to deforestation and degradation. Despite the world's government's commitment to reduce or end forest loss, the trend is still increasing, specifically in the tropics, which holds 45% of the existing forest and nearly 50% of the Earth's biodiversity. The conversion of forests for food crops, commodity crops, mining, charcoal, and the need for building materials has been identified as the main leading driver of forest loss. A growing world population and increasing demand for agricultural products, especially those of commodity crops, are expected to increase demand for agricultural land and put additional pressure on forests, while changing climate patterns will aggravate this by affecting food production. Several actions, including improving farming practices (e.g. intensification), and buying only sustainable agricultural products, have been proposed to reduce pressure on forests. However, before any action is taken, the assessment of the origin, location and extent of forest loss and what drives it need to be identified.

show abstract

Section: Discussionsupporting

confidence: 81%

Assessing land use following tropical deforestation : Combining remote sensing and deep learning

Masolele

View full text Add to dashboard Cite

show abstract

Mapping the diversity of land uses following deforestation across Africa

Masolele,

Marcos,

De Sy

et al. 2024

Sci Rep

View full text Add to dashboard Cite

African forest are increasingly in decline as a result of land-use conversion due to human activities. However, a consistent and detailed characterization and mapping of land-use change that results in forest loss is not available at the spatial-temporal resolution and thematic levels suitable for decision-making at the local and regional scales; so far they have only been provided on coarser scales and restricted to humid forests. Here we present the first high-resolution (5 m) and continental-scale mapping of land use following deforestation in Africa, which covers an estimated 13.85% of the global forest area, including humid and dry forests. We use reference data for 15 different land-use types from 30 countries and implement an active learning framework to train a deep learning model for predicting land-use following deforestation with an F1-score of $$84\pm 0.7$$ 84 ± 0.7 for the whole of Africa. Our results show that the causes of forest loss vary by region. In general, small-scale cropland is the dominant driver of forest loss in Africa, with hotspots in Madagascar and DRC. In addition, commodity crops such as cacao, oil palm, and rubber are the dominant drivers of forest loss in the humid forests of western and central Africa, forming an “arc of commodity crops” in that region. At the same time, the hotspots for cashew are found to increasingly dominate in the dry forests of both western and south-eastern Africa, while larger hotspots for large-scale croplands were found in Nigeria and Zambia. The increased expansion of cacao, cashew, oil palm, rubber, and large-scale croplands observed in humid and dry forests of western and south-eastern Africa suggests they are vulnerable to future land-use changes by commodity crops, thus creating challenges for achieving the zero deforestation supply chains, support REDD+ initiatives, and towards sustainable development goals.

show abstract

Blending Query Strategy of Active Learning for Imbalanced Data

Kim

Yoo

2022

IEEE Access

View full text Add to dashboard Cite

When the data is imbalanced, often observed in the real-world, important minor class instances that are conducive to accurately predicting the decision boundary are less likely to be queried in the active learning for classification task. Therefore, mitigating the effect of the imbalance in learning is necessary for achieving better generalization. For the alleviation of this problem, this paper considers an active learning algorithm referred to as the blending minority preferential (BMP). The BMP systematically adapts to blend conventional query strategies with the proposed minority preferential queries on a randomly sampled pool dataset. The proposed minority preferential queries are conditioned on an unlabeled data instance predicted to belong to the minor class. A multi-armed bandit is involved in the blending of the two different types of queries for achieving high accuracy and balanced learning in obtaining each query set. The performance of the BMP is validated on datasets that include three imbalanced datasets having binary labels, eleven small structured datasets, modified Fashion-MNIST, CIFAR10 dataets, and two real datasets of the PlantSC and HAM10000. The comparison result shows that the BMP can mitigate imbalanced learning and achieves higher accuracy combined with higher G-mean, compared with conventional queries or other query-based imbalanced active learning.

show abstract

Applying active learning strategy to classify large scale data with imbalanced classes

Cited by 4 publications

References 11 publications

Assessing land use following tropical deforestation : Combining remote sensing and deep learning

Assessing land use following tropical deforestation : Combining remote sensing and deep learning

Mapping the diversity of land uses following deforestation across Africa

Blending Query Strategy of Active Learning for Imbalanced Data

Contact Info

Product

Resources

About