The Impact of Imbalanced Training Data on Local Matching Learning of Ontologies

In this paper, we propose a new method for code-switching (CS) automatic speech recognition (ASR) in Korean. First, the phonetic variations in English pronunciation spoken by Korean speakers should be considered. Thus, we tried to find a unified pronunciation model based on phonetic knowledge and deep learning. Second, we extracted the CS sentences semantically similar to the target domain and then applied the language model (LM) adaptation to solve the biased modeling toward Korean due to the imbalanced training data. In this experiment, training data were AI Hub (1033 h) in Korean and Librispeech (960 h) in English. As a result, when compared to the baseline, the proposed method improved the error reduction rate (ERR) by up to 11.6% with phonetic variant modeling and by 17.3% when semantically similar sentences were applied to the LM adaptation. If we considered only English words, the word correction rate improved up to 24.2% compared to that of the baseline. The proposed method seems to be very effective in CS speech recognition.

Download Full-text

Copy-Move Forgery Detection (CMFD) Using Deep Learning for Image and Video Forensics

Journal of Imaging ◽

10.3390/jimaging7030059 ◽

2021 ◽

Vol 7 (3) ◽

pp. 59

Author(s):

Yohanna Rodriguez-Ortega ◽

Dora M. Ballesteros ◽

Diego Renza

Keyword(s):

Deep Learning ◽

Transfer Learning ◽

Training Data ◽

Video Editing ◽

Forgery Detection ◽

Copy Move Forgery Detection ◽

The Impact ◽

And Training ◽

Selection Of ◽

Traditional Image

With the exponential growth of high-quality fake images in social networks and media, it is necessary to develop recognition algorithms for this type of content. One of the most common types of image and video editing consists of duplicating areas of the image, known as the copy-move technique. Traditional image processing approaches manually look for patterns related to the duplicated content, limiting their use in mass data classification. In contrast, approaches based on deep learning have shown better performance and promising results, but they present generalization problems with a high dependence on training data and the need for appropriate selection of hyperparameters. To overcome this, we propose two approaches that use deep learning, a model by a custom architecture and a model by transfer learning. In each case, the impact of the depth of the network is analyzed in terms of precision (P), recall (R) and F1 score. Additionally, the problem of generalization is addressed with images from eight different open access datasets. Finally, the models are compared in terms of evaluation metrics, and training and inference times. The model by transfer learning of VGG-16 achieves metrics about 10% higher than the model by a custom architecture, however, it requires approximately twice as much inference time as the latter.

Download Full-text

Underwater Acoustic Target Recognition Based on Generative Adversarial Network Data Augmentation

INTER-NOISE and NOISE-CON Congress and Conference Proceedings ◽

10.3397/in-2021-2737 ◽

2021 ◽

Vol 263 (2) ◽

pp. 4558-4564

Author(s):

Minghong Zhang ◽

Xinwei Luo

Keyword(s):

Data Augmentation ◽

Target Recognition ◽

Training Data ◽

Small Samples ◽

Generative Adversarial Network ◽

Data Set ◽

Underwater Acoustic ◽

Adversarial Network ◽

Acoustic Target ◽

The Impact

Underwater acoustic target recognition is an important aspect of underwater acoustic research. In recent years, machine learning has been developed continuously, which is widely and effectively applied in underwater acoustic target recognition. In order to acquire good recognition results and reduce the problem of overfitting, Adequate data sets are essential. However, underwater acoustic samples are relatively rare, which has a certain impact on recognition accuracy. In this paper, in addition of the traditional audio data augmentation method, a new method of data augmentation using generative adversarial network is proposed, which uses generator and discriminator to learn the characteristics of underwater acoustic samples, so as to generate reliable underwater acoustic signals to expand the training data set. The expanded data set is input into the deep neural network, and the transfer learning method is applied to further reduce the impact caused by small samples by fixing part of the pre-trained parameters. The experimental results show that the recognition result of this method is better than the general underwater acoustic recognition method, and the effectiveness of this method is verified.

Download Full-text

Efficient Patch-Wise Semantic Segmentation for Large-Scale Remote Sensing Images

Sensors ◽

10.3390/s18103232 ◽

2018 ◽

Vol 18 (10) ◽

pp. 3232 ◽

Cited By ~ 17

Author(s):

Yan Liu ◽

Qirui Ren ◽

Jiahui Geng ◽

Meng Ding ◽

Jiangyun Li

Keyword(s):

Remote Sensing ◽

High Resolution ◽

Large Scale ◽

Semantic Segmentation ◽

Remote Sensing Image ◽

Training Data ◽

Land Resources ◽

Remote Sensing Images ◽

Training Strategy ◽

The Impact

Efficient and accurate semantic segmentation is the key technique for automatic remote sensing image analysis. While there have been many segmentation methods based on traditional hand-craft feature extractors, it is still challenging to process high-resolution and large-scale remote sensing images. In this work, a novel patch-wise semantic segmentation method with a new training strategy based on fully convolutional networks is presented to segment common land resources. First, to handle the high-resolution image, the images are split as local patches and then a patch-wise network is built. Second, training data is preprocessed in several ways to meet the specific characteristics of remote sensing images, i.e., color imbalance, object rotation variations and lens distortion. Third, a multi-scale training strategy is developed to solve the severe scale variation problem. In addition, the impact of conditional random field (CRF) is studied to improve the precision. The proposed method was evaluated on a dataset collected from a capital city in West China with the Gaofen-2 satellite. The dataset contains ten common land resources (Grassland, Road, etc.). The experimental results show that the proposed algorithm achieves 54.96% in terms of mean intersection over union (MIoU) and outperforms other state-of-the-art methods in remote sensing image segmentation.

Download Full-text

DANNP: an efficient artificial neural network pruning tool

PeerJ Computer Science ◽

10.7717/peerj-cs.137 ◽

2017 ◽

Vol 3 ◽

pp. e137 ◽

Cited By ~ 7

Author(s):

Mona Alshahrani ◽

Othman Soufan ◽

Arturo Magana-Mora ◽

Vladimir B. Bajic

Keyword(s):

Neural Network ◽

State Of The Art ◽

Model Performance ◽

Training Data ◽

Classification Problems ◽

Link Type ◽

On Line ◽

Pruning Algorithms ◽

Artificial Neural ◽

The Impact

Background Artificial neural networks (ANNs) are a robust class of machine learning models and are a frequent choice for solving classification problems. However, determining the structure of the ANNs is not trivial as a large number of weights (connection links) may lead to overfitting the training data. Although several ANN pruning algorithms have been proposed for the simplification of ANNs, these algorithms are not able to efficiently cope with intricate ANN structures required for complex classification problems. Methods We developed DANNP, a web-based tool, that implements parallelized versions of several ANN pruning algorithms. The DANNP tool uses a modified version of the Fast Compressed Neural Network software implemented in C++ to considerably enhance the running time of the ANN pruning algorithms we implemented. In addition to the performance evaluation of the pruned ANNs, we systematically compared the set of features that remained in the pruned ANN with those obtained by different state-of-the-art feature selection (FS) methods. Results Although the ANN pruning algorithms are not entirely parallelizable, DANNP was able to speed up the ANN pruning up to eight times on a 32-core machine, compared to the serial implementations. To assess the impact of the ANN pruning by DANNP tool, we used 16 datasets from different domains. In eight out of the 16 datasets, DANNP significantly reduced the number of weights by 70%–99%, while maintaining a competitive or better model performance compared to the unpruned ANN. Finally, we used a naïve Bayes classifier derived with the features selected as a byproduct of the ANN pruning and demonstrated that its accuracy is comparable to those obtained by the classifiers trained with the features selected by several state-of-the-art FS methods. The FS ranking methodology proposed in this study allows the users to identify the most discriminant features of the problem at hand. To the best of our knowledge, DANNP (publicly available at www.cbrc.kaust.edu.sa/dannp) is the only available and on-line accessible tool that provides multiple parallelized ANN pruning options. Datasets and DANNP code can be obtained at www.cbrc.kaust.edu.sa/dannp/data.php and https://doi.org/10.5281/zenodo.1001086.

Download Full-text

Estimating probability of banking crises using random forest

IAES International Journal of Artificial Intelligence (IJ-AI) ◽

10.11591/ijai.v10.i2.pp407-413 ◽

2021 ◽

Vol 10 (2) ◽

pp. 407

Author(s):

Sri Hartini ◽

Zuherman Rustam ◽

Glori Stephani Saragih ◽

María Jesús Segovia Vargas

Keyword(s):

Random Forest ◽

State Of The Art ◽

Banking Crises ◽

Training Data ◽

Annual Data ◽

Systemic Crisis ◽

Classification And Regression ◽

Systemic Crises ◽

The Impact ◽

Better Than

<span id="docs-internal-guid-4935b5ce-7fff-d9fa-75c7-0c6a5aa1f9a6"><span>Banks have a crucial role in the financial system. When many banks suffer from the crisis, it can lead to financial instability. According to the impact of the crises, the banking crisis can be divided into two categories, namely systemic and non-systemic crisis. When systemic crises happen, it may cause even stable banks bankrupt. Hence, this paper proposed a random forest for estimating the probability of banking crises as prevention action. Random forest is well-known as a robust technique both in classification and regression, which is far from the intervention of outliers and overfitting. The experiments were then constructed using the financial crisis database, containing a sample of 79 countries in the period 1981-1999 (annual data). This dataset has 521 samples consisting of 164 crisis samples and 357 non-crisis cases. From the experiments, it was concluded that utilizing 90 percent of training data would deliver 0.98 accuracy, 0.92 sensitivity, 1.00 precision, and 0.96 F1-Score as the highest score than other percentages of training data. These results are also better than state-of-the-art methods used in the same dataset. Therefore, the proposed method is shown promising results to predict the probability of banking crises.</span></span>

Download Full-text

ASSESSING LIDAR TRAINING DATA QUANTITIES FOR CLASSIFICATION MODELS

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xlvi-4-w4-2021-101-2021 ◽

2021 ◽

Vol XLVI-4/W4-2021 ◽

pp. 101-106

Author(s):

O. Majgaonkar ◽

K. Panchal ◽

D. Laefer ◽

M. Stanley ◽

Y. Zaki

Keyword(s):

Point Cloud ◽

Classification Performance ◽

Training Data ◽

Lidar Data ◽

Data Sets ◽

Classification Models ◽

Fundamental Properties ◽

Point Cloud Classification ◽

The Impact ◽

Insight Into

Abstract. Classifying objects within aerial Light Detection and Ranging (LiDAR) data is an essential task to which machine learning (ML) is applied increasingly. ML has been shown to be more effective on LiDAR than imagery for classification, but most efforts have focused on imagery because of the challenges presented by LiDAR data. LiDAR datasets are of higher dimensionality, discontinuous, heterogenous, spatially incomplete, and often scarce. As such, there has been little examination into the fundamental properties of the training data required for acceptable performance of classification models tailored for LiDAR data. The quantity of training data is one such crucial property, because training on different sizes of data provides insight into a model’s performance with differing data sets. This paper assesses the impact of training data size on the accuracy of PointNet, a widely used ML approach for point cloud classification. Subsets of ModelNet ranging from 40 to 9,843 objects were validated on a test set of 400 objects. Accuracy improved logarithmically; decelerating from 45 objects onwards, it slowed significantly at a training size of 2,000 objects, corresponding to 20,000,000 points. This work contributes to the theoretical foundation for development of LiDAR-focused models by establishing a learning curve, suggesting the minimum quantity of manually labelled data necessary for satisfactory classification performance and providing a path for further analysis of the effects of modifying training data characteristics.

Download Full-text

Unsupervised learning of part-of-speech guessing rules

Natural Language Engineering ◽

10.1017/s1351324996001271 ◽

1996 ◽

Vol 2 (2) ◽

pp. 111-136 ◽

Cited By ~ 1

Author(s):

ANDREI MIKHEEV

Keyword(s):

Training Data ◽

Evaluation Methodology ◽

Parts Of Speech ◽

Part Of Speech Tagging ◽

Part Of Speech ◽

Substantial Problem ◽

Rule Sets ◽

Induction Experiment ◽

Unknown Words ◽

The Impact

Words unknown to the lexicon present a substantial problem to part-of-speech tagging. In this paper we present a technique for fully unsupervised acquisition of rules which guess possible parts of speech for unknown words. This technique does not require specially prepared training data, and uses instead the lexicon supplied with a tagger and word frequencies collected from a raw corpus. Three complimentary sets of word-guessing rules are statistically induced: prefix morphological rules, suffix morphological rules and ending guessing rules. The acquisition process is strongly associated with guessing-rule evaluation methodology which is solely dedicated to the performance of part-of-speech guessers. Using the proposed technique a guessing-rule induction experiment was performed on the Brown Corpus data and rule-sets, with a highly competitive performance, were produced and compared with the state-of-the-art. To evaluate the impact of the word-guessing component on the overall tagging performance, it was integrated into a stochastic and a rule-based tagger and applied to texts with unknown words.

Download Full-text

The Evolving Role of Task-Group Leaders, A Field Study

Proceedings of the Human Factors Society Annual Meeting ◽

10.1518/107118188786762333 ◽

1988 ◽

Vol 32 (13) ◽

pp. 760-764

Author(s):

Robert F. Randolph

Keyword(s):

Group Performance ◽

Training Data ◽

Underground Coal Mining ◽

Group Leaders ◽

Underground Coal Mines ◽

And Performance ◽

Task Oriented ◽

The Impact ◽

And Training

Leaders of task-oriented production groups play an important role in their group's functioning and performance. That role also evolves as groups mature and learn to work together more smoothly. The present study uses a functional analysis of the evolving role of supervisors of underground coal mining crews to evaluate the impact of supervisors' characteristics and behaviors on their crews' efficiency and safety, and makes recommendations for improving supervisory selection and training. Data were gathered from a sample of 138 supervisors at 13 underground coal mines. Detailed structured observations of the supervisors indicated that most of their time was spent attending to hardware and paperwork, while comparatively little time was spent on person to person “leadership”. The findings point out that while group needs changed over time, the supervisors' behaviors typically did not keep pace and probably restricted group performance.

Download Full-text