A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf
.
Filters
Towards A Holistic View of Bias in Machine Learning: Bridging Algorithmic Fairness and Imbalanced Learning
[article]
2022
arXiv
pre-print
A key element in achieving algorithmic fairness with respect to protected groups is the simultaneous reduction of class and protected group imbalance in the underlying training data, which facilitates ...
We discuss the importance of bridging imbalanced learning and group fairness by showing how key concepts in these fields overlap and complement each other; and propose a novel oversampling algorithm, Fair ...
We show that common approach used in imbalanced learning -data oversampling -can be used to increase model fairness and accuracy. Main contributions. ...
arXiv:2207.06084v1
fatcat:wyl2wkdmmbfwjgqfgq2okbpbsu
A Multitask Convolutional Neural Network for Artwork Appreciation
2022
Mobile Information Systems
networks to complete artwork art appreciation, and an oversampling method and multitask learning method are used to improve the overall recognition accuracy. (3) Compared with the combination of traditional ...
. (2) Based on the artwork art appreciation dataset, an AlexNet-based convolutional neural network model is proposed to utilize the powerful feature extraction and classification capabilities of neural ...
Acknowledgments This work is supported by the Dalian Polytechnic University. ...
doi:10.1155/2022/8804711
fatcat:kiz56kb5tncafaywrg2tkl3d4u
Exploring Social Relationships in Text Streams
2016
EAI Endorsed Transactions on Scalable Information Systems
In this paper, we specify the research gap and present a review report for dynamic text streams. ...
Most raw information related to social relationships are continuously generated by social networks in a form of text, for the reason that it has the lowest storage consumption while still possesses powerful ...
The traditional methods for classification need to infer a static judgement function from the labeled training data set so that it can be further used to classify new data sets. ...
doi:10.4108/eai.9-8-2016.151631
fatcat:oacil33o55fpdla722dyanjjoe
Imbalanced Dataset Effect on CNN-Based Classifier Performance for Face Recognition
2024
International Journal of Artificial Intelligence & Applications
Initially, we crafted a Convolutional Neural Network model for facial recognition, integrating hybrid resampling methods (oversampling and under-sampling) to address dataset imbalances. ...
This study underscores the efficacy of data resampling approaches in augmenting the performance of Face Recognition models, presenting prospects for more dependable and efficient future systems. ...
In our model, we employed a combined strategy of oversampling and under-sampling to generate a balanced dataset (Hybrid method). ...
doi:10.5121/ijaia.2024.15102
fatcat:ayk7wkrczjfv3oj2zp3me3brdi
Diversified RACE Sampling on Data Streams Applied to Metagenomic Sequence Analysis
[article]
2019
bioRxiv
pre-print
While there are many methods to identify common elements in data streams efficiently, fast and memory-efficient diversity sampling remains a challenging and fundamental data streaming problem with few ...
In theory, the best diversity sampling methods are based on a simple greedy algorithm that compares the current sequence with a large pool of sampled sequences and decides whether to accept or reject the ...
In spite of recent methods designed to handle the data deluge [25] , new methods are needed that focus on data compression, sampling, and summarization to facilitate faster algorithms and practical systems ...
doi:10.1101/852889
fatcat:zlez2dhwdjei3afpjbdoj7rs5m
Wireless AI-Powered IoT Sensors for Laboratory Mice Behavior Recognition
[article]
2020
bioRxiv
pre-print
A combination of oversampling and undersampling is used to handle imbalanced classes, and feature selection provides the optimal number of features. ...
The system collects motion data (i.e., three axes linear accelerations and three axes angular velocities) from the IoT sensors attached to different mice, and classifies these data into different behaviors ...
Random sampling is a typical method to solve the imbalance by adjusting the class distribution of a data set. Two kinds of random sampling methods can be applied: oversampling and undersampling. ...
doi:10.1101/2020.07.23.217190
fatcat:yk6knd6axvhxfcer7h6xscfaka
Machine Learning in Detecting COVID-19 Misinformation on Twitter
2021
Future Internet
In addition to the positives of the convenience of these platforms, they are often used for propagating malicious data or information. ...
The proposed models are long short-term memory (LSTM) networks, which is a special type of RNN; a multichannel convolutional neural network (MC-CNN); and k-nearest neighbors (KNN). ...
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest. ...
doi:10.3390/fi13100244
fatcat:trvqzlik6ffcvfhkhnqsopokz4
A Comparative Study of Machine Learning Methods for Computational Modeling of the Selective Laser Melting Additive Manufacturing Process
2022
Applied Sciences
Of the three algorithms, the PSO performed well, and the DNN model was found to be the most efficient surrogate model compared to the PCE. ...
A surrogate model was constructed with a deep neural network (DNN) or polynomial chaos expansion (PCE) to generate a response surface between the SLM output and the input variables. ...
the input data in the offline stage, which allows a surrogate model to be obtained using a regression method on a reduced basis; (2) Predictions on new data performed on the online stage using the surrogate ...
doi:10.3390/app12052324
fatcat:qv2trqhhlvfslnglxqfkovzjia
Rich Feature Combination for Cost-Based Broad Learning System
2019
IEEE Access
As an alternative approach of learning in deep structure, the BLS develops an incremental learning neural network that can be modeled in a flexible way, and becomes a promising technique in the field of ...
To further improve the performance of BLS, our focus is to investigate these algorithms which can enhance the BLS. ...
Imbalanced distribution thus presents a new challenge to the BLS community. To bridge the gap between BLS and imbalanced learning, the second topic of our study is cost-sensitive BLS. ...
doi:10.1109/access.2018.2885164
fatcat:lb2smwbwsbhhjomrfovn6nc7sq
Maintenance intervention predictions using entity-embedding neural networks
2020
Automation in Construction
A B S T R A C T Data-driven decision support can substantially aid in smart and efficient maintenance planning of road bridges. ...
The results of models are further evaluated by instance-level explanations, which provide insights about essential features and explain the importance of data attributes for a particular task. ...
Acknowledgement This study has been performed under funding from the European Union's Horizon 2020 -Research and Innovation Framework Programme with grant agreement No 636285 DESTination Rail. ...
doi:10.1016/j.autcon.2020.103202
fatcat:jhxa5ir2a5dw5ncerjbxio2om4
Towards Explainable, Safe Autonomous Driving with Language Embeddings for Novelty Identification and Active Learning: Framework and Experimental Analysis with Real-World Data Sets
[article]
2024
arXiv
pre-print
Our results demonstrate the effectiveness of language-driven embeddings in identifying novel elements and generating explanations of data, and we further discuss potential applications in safe takeovers ...
From the generated clusters, we further present methods for generating textual explanations of elements which differentiate scenes classified as novel from other scenes in the data pool, presenting qualitative ...
The authors would also like to acknowledge the support of Qualcomm through the Qualcomm Innovation Fellowship, and thank mentors for their valuable feedback. ...
arXiv:2402.07320v1
fatcat:xzzoeqawbnhr3h2pmmii6udsrm
A Novel Incremental Cross-Modal Hashing Approach
[article]
2020
arXiv
pre-print
In this work, we propose a novel incremental cross-modal hashing algorithm termed "iCMH", which can adapt itself to handle incoming data of new categories. ...
At every stage, a small amount of old category data termed "exemplars" is is used so as not to forget the old data while trying to learn for the new incoming data, i.e. to avoid catastrophic forgetting ...
Incremental methods for classification to handle data from new category or new task is a well studied problem and seminal approaches have been proposed in [22] [8] [18] [5] [27] [28] . ...
arXiv:2002.00677v1
fatcat:5o24pnenr5d5ni5eeydxws6uym
Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments
2016
PLoS ONE
Coping with scarcity of labeled data is a common problem in sound classification tasks. ...
Approaches for classifying sounds are commonly based on supervised learning algorithms, which require labeled data which is often scarce and leads to models that do not generalize well. ...
Acknowledgments The authors would like to thank Zixing Zhang for his help with collecting data, and Jun Deng for their valuable feedback on the study design and on earlier versions of this manuscript. ...
doi:10.1371/journal.pone.0162075
pmid:27627768
pmcid:PMC5023122
fatcat:a3cerzf6s5exlbfktgqore7xue
Damage Detection in Structures by Using Imbalanced Classification Algorithms
2024
Mathematics
Four imbalanced classification algorithms are applied to two benchmark structures: the first, a numerical model of a four-story steel building, and the second, a bridge constructed in China. ...
While many existing methods prove effective when the number of data points in both healthy and damaged states is equal, this article employs algorithms tailored for detecting damage in situations where ...
One-Class Support Vector Machines The main idea of SVMs is to map the input data points to a high-dimensional feature space and find a hyperplane, and the algorithm is chosen to maximize the distance from ...
doi:10.3390/math12030432
fatcat:qewoifuexnhkzg7ur7t7jilmoq
Undersampling Strategy for Machine-learned Deterioration Regression Model in Concrete Bridges
2020
Journal of advanced concrete technology
This study applies machine learning to a regression model of the crack damage grade in concrete bridges, using imbalanced inspection data. ...
Inspection data of actual concrete structures should be analyzed to elucidate the deterioration mechanism and construct a regression model. ...
Acknowledgements This work was supported by Ministry of Land, Infrastructure, Transport and Tourism (MLIT), Construction Technology Research and Development Grant Program(CRD), the Cabinet Office, Government ...
doi:10.3151/jact.18.753
fatcat:bgss76kj5vdobm5pyti43hslk4
« Previous
Showing results 1 — 15 out of 1,003 results