Multimedia Feature Mapping and Correlation Learning for Cross-Modal Retrieval

This article describes how with the rapid increasing of multimedia content on the Internet, the need for effective cross-modal retrieval has attracted much attention recently. Many related works ignore the latent semantic correlations of modalities in the non-linear space and the extraction of high-level modality features, which only focuses on the semantic mapping of modalities in linear space and the use of low-level artificial features as modality feature representation. To solve these issues, the authors first utilizes convolutional neural networks and topic modal to obtain a high-level semantic feature of various modalities. Sequentially, they propose a supervised learning algorithm based on a kernel with partial least squares that can capture semantic correlations across modalities. Finally, the joint model of different modalities is learnt by the training set. Extensive experiments are conducted on three benchmark datasets that include Wikipedia, Pascal and MIRFlickr. The results show that the proposed approach achieves better retrieval performance over several state-of-the-art approaches.

Download Full-text

Broccoli: combining phylogenetic and network analyses for orthology assignment

10.1101/2019.12.13.875831 ◽

2019 ◽

Cited By ~ 1

Author(s):

Romain Derelle ◽

Hervé Philippe ◽

John K. Colbourne

Keyword(s):

Learning Algorithm ◽

Phylogenetic Analyses ◽

Gene Clustering ◽

Comparative Genomic ◽

Orthology Assignment ◽

Bioinformatic Tools ◽

Network Analyses ◽

Benchmark Datasets ◽

Genomic Studies ◽

High Level

AbstractOrthology assignment is a key step of comparative genomic studies, for which many bioinformatic tools have been developed. However, all gene clustering pipelines are based on the analysis of protein distances, which are subject to many artefacts. In this paper we introduce Broccoli, a user-friendly pipeline designed to infer, with high precision, orthologous groups and pairs of proteins using a phylogeny-based approach. Briefly, Broccoli performs ultra-fast phylogenetic analyses on most proteins and builds a network of orthologous relationships. Orthologous groups are then identified from the network using a parameter-free machine learning algorithm. Broccoli is also able to detect chimeric proteins resulting from gene-fusion events and to assign these proteins to the corresponding orthologous groups. Tested on two benchmark datasets, Broccoli outperforms current orthology pipelines. In addition, Broccoli is scalable, with runtimes similar to those of recent distance-based pipelines. Given its high level of performance and efficiency, this new pipeline represents a suitable choice for comparative genomic studies.Broccoli is freely available at https://github.com/rderelle/Broccoli.

Download Full-text

iMethyl-Deep: N6 Methyladenosine Identification of Yeast Genome with Automatic Feature Extraction Technique by Using Deep Learning Algorithm

Genes ◽

10.3390/genes11050529 ◽

2020 ◽

Vol 11 (5) ◽

pp. 529 ◽

Cited By ~ 8

Author(s):

Omid Mahmoudi ◽

Abdul Wahab ◽

Kil To Chong

Keyword(s):

Learning Algorithm ◽

Feature Representation ◽

Yeast Genome ◽

Single Nucleotide ◽

Automatic Feature Extraction ◽

Deep Learning Algorithm ◽

Wide Range ◽

Benchmark Datasets ◽

Nucleotide Resolution ◽

Single Nucleotide Resolution

One of the most common and well studied post-transcription modifications in RNAs is N6-methyladenosine (m6A) which has been involved with a wide range of biological processes. Over the past decades, N6-methyladenosine produced some positive consequences through the high-throughput laboratory techniques but still, these lab processes are time consuming and costly. Diverse computational methods have been proposed to identify m6A sites accurately. In this paper, we proposed a computational model named iMethyl-deep to identify m6A Saccharomyces Cerevisiae on two benchmark datasets M6A2614 and M6A6540 by using single nucleotide resolution to convert RNA sequence into a high quality feature representation. The iMethyl-deep obtained 89.19% and 87.44% of accuracy on M6A2614 and M6A6540 respectively which show that our proposed method outperforms the state-of-the-art predictors, at least 8.44%, 8.96%, 8.69% and 0.173 on M6A2614 and 15.47%, 28.52%, 25.54 and 0.5 on M6A6540 higher in terms of four metrics Sp, Sn, ACC and MCC respectively. Meanwhile, M6A6540 dataset never used to train a model.

Download Full-text

Earthquake-Damaged Buildings Detection in Very High-Resolution Remote Sensing Images Based on Object Context and Boundary Enhanced Loss

Remote Sensing ◽

10.3390/rs13163119 ◽

2021 ◽

Vol 13 (16) ◽

pp. 3119

Author(s):

Chao Wang ◽

Xing Qiu ◽

Hai Huan ◽

Shuai Wang ◽

Yan Zhang ◽

...

Keyword(s):

Remote Sensing ◽

High Resolution ◽

Feature Representation ◽

Detection Accuracy ◽

Remote Sensing Images ◽

Convolutional Networks ◽

Fully Convolutional Networks ◽

Benchmark Datasets ◽

High Level ◽

Very High

Fully convolutional networks (FCN) such as UNet and DeepLabv3+ are highly competitive when being applied in the detection of earthquake-damaged buildings in very high-resolution (VHR) remote sensing images. However, existing methods show some drawbacks, including incomplete extraction of different sizes of buildings and inaccurate boundary prediction. It is attributed to a deficiency in the global context-aware and inaccurate correlation mining in the spatial context as well as failure to consider the relative positional relationship between pixels and boundaries. Hence, a detection method for earthquake-damaged buildings based on the object contextual representations (OCR) and boundary enhanced loss (BE loss) was proposed. At first, the OCR module was separately embedded into high-level feature extractions of the two networks DeepLabv3+ and UNet in order to enhance the feature representation; in addition, a novel loss function, that is, BE loss, was designed according to the distance between the pixels and boundaries to force the networks to pay more attention to the learning of the boundary pixels. Finally, two improved networks (including OB-DeepLabv3+ and OB-UNet) were established according to the two strategies. To verify the performance of the proposed method, two benchmark datasets (including YSH and HTI) for detecting earthquake-damaged buildings were constructed according to the post-earthquake images in China and Haiti in 2010, respectively. The experimental results show that both the embedment of the OCR module and application of BE loss contribute to significantly increasing the detection accuracy of earthquake-damaged buildings and the two proposed networks are feasible and effective.

Download Full-text

Prototype-Based Discriminative Feature Representation for Class-incremental Cross-modal Retrieval

International Journal of Pattern Recognition and Artificial Intelligence ◽

10.1142/s021800142150018x ◽

2020 ◽

pp. 2150018

Author(s):

Shaoquan Zhu ◽

Yong Feng ◽

Mingliang Zhou ◽

Baohua Qiang ◽

Bin Fang ◽

...

Keyword(s):

Learning Algorithm ◽

Feature Learning ◽

Feature Representation ◽

Retrieval Performance ◽

Retrieval Task ◽

Semantic Class ◽

Common Space ◽

Open World ◽

Discriminative Feature ◽

Prototype Learning

Cross-modal retrieval aims to retrieve the related items from various modalities with respect to a query from any type. The key challenge of cross-modal retrieval is to learn more discriminative representations between different category, as well as expand to an unseen class retrieval in the open world retrieval task. To tackle the above problem, in this paper, we propose a prototype learning-based discriminative feature learning (PLDFL) to learn more discriminative representations in a common space. First, we utilize a prototype learning algorithm to cluster these samples labeled with the same semantic class, by jointly taking into consideration the intra-class compactness and inter-class sparsity without discriminative treatments. Second, we use the weight-sharing strategy to model the correlations of cross-modal samples to narrow down the modality gap. Finally, we apply the prototype to achieve class-incremental learning to prove the robustness of our proposed approach. According to our experimental results, significant retrieval performance in terms of mAP can be achieved on average compared to several state-of-the-art approaches.

Download Full-text

Towards Computational Models of Identifying Protein Ubiquitination Sites

Current Drug Targets ◽

10.2174/1389450119666180924150202 ◽

2019 ◽

Vol 20 (5) ◽

pp. 565-578 ◽

Cited By ~ 1

Author(s):

Lidong Wang ◽

Ruijun Zhang

Keyword(s):

Computational Methods ◽

Computational Models ◽

Feature Representation ◽

Biological Sequence ◽

Post Translational Modification ◽

Test Dataset ◽

Protein Ubiquitination ◽

Protein Functions ◽

Independent Test Dataset ◽

Benchmark Datasets

Ubiquitination is an important post-translational modification (PTM) process for the regulation of protein functions, which is associated with cancer, cardiovascular and other diseases. Recent initiatives have focused on the detection of potential ubiquitination sites with the aid of physicochemical test approaches in conjunction with the application of computational methods. The identification of ubiquitination sites using laboratory tests is especially susceptible to the temporality and reversibility of the ubiquitination processes, and is also costly and time-consuming. It has been demonstrated that computational methods are effective in extracting potential rules or inferences from biological sequence collections. Up to the present, the computational strategy has been one of the critical research approaches that have been applied for the identification of ubiquitination sites, and currently, there are numerous state-of-the-art computational methods that have been developed from machine learning and statistical analysis to undertake such work. In the present study, the construction of benchmark datasets is summarized, together with feature representation methods, feature selection approaches and the classifiers involved in several previous publications. In an attempt to explore pertinent development trends for the identification of ubiquitination sites, an independent test dataset was constructed and the predicting results obtained from five prediction tools are reported here, together with some related discussions.

Download Full-text

An IoT-Focused Intrusion Detection System Approach Based on Preprocessing Characterization for Cybersecurity Datasets

Sensors ◽

10.3390/s21020656 ◽

2021 ◽

Vol 21 (2) ◽

pp. 656

Author(s):

Xavier Larriva-Novo ◽

Víctor A. Villagrá ◽

Mario Vega-Barbas ◽

Diego Rivera ◽

Mario Sanz Rodrigo

Keyword(s):

Machine Learning ◽

Intrusion Detection ◽

High Performance ◽

Learning Algorithm ◽

Detection System ◽

Machine Learning Algorithms ◽

Statistical Characteristics ◽

Detection Techniques ◽

Traffic Characteristics ◽

Benchmark Datasets

Security in IoT networks is currently mandatory, due to the high amount of data that has to be handled. These systems are vulnerable to several cybersecurity attacks, which are increasing in number and sophistication. Due to this reason, new intrusion detection techniques have to be developed, being as accurate as possible for these scenarios. Intrusion detection systems based on machine learning algorithms have already shown a high performance in terms of accuracy. This research proposes the study and evaluation of several preprocessing techniques based on traffic categorization for a machine learning neural network algorithm. This research uses for its evaluation two benchmark datasets, namely UGR16 and the UNSW-NB15, and one of the most used datasets, KDD99. The preprocessing techniques were evaluated in accordance with scalar and normalization functions. All of these preprocessing models were applied through different sets of characteristics based on a categorization composed by four groups of features: basic connection features, content characteristics, statistical characteristics and finally, a group which is composed by traffic-based features and connection direction-based traffic characteristics. The objective of this research is to evaluate this categorization by using various data preprocessing techniques to obtain the most accurate model. Our proposal shows that, by applying the categorization of network traffic and several preprocessing techniques, the accuracy can be enhanced by up to 45%. The preprocessing of a specific group of characteristics allows for greater accuracy, allowing the machine learning algorithm to correctly classify these parameters related to possible attacks.

Download Full-text

UAV Image Multi-Labeling with Data-Efficient Transformers

Applied Sciences ◽

10.3390/app11093974 ◽

2021 ◽

Vol 11 (9) ◽

pp. 3974

Author(s):

Laila Bashmal ◽

Yakoub Bazi ◽

Mohamad Mahmoud Al Rahhal ◽

Haikel Alhichri ◽

Naif Al Ajlan

Keyword(s):

Data Augmentation ◽

Feature Representation ◽

Aerial Image ◽

Remote Sensing Images ◽

Training Set ◽

Proposed Model ◽

Class Labels ◽

Using Data ◽

Uav Image

In this paper, we present an approach for the multi-label classification of remote sensing images based on data-efficient transformers. During the training phase, we generated a second view for each image from the training set using data augmentation. Then, both the image and its augmented version were reshaped into a sequence of flattened patches and then fed to the transformer encoder. The latter extracts a compact feature representation from each image with the help of a self-attention mechanism, which can handle the global dependencies between different regions of the high-resolution aerial image. On the top of the encoder, we mounted two classifiers, a token and a distiller classifier. During training, we minimized a global loss consisting of two terms, each corresponding to one of the two classifiers. In the test phase, we considered the average of the two classifiers as the final class labels. Experiments on two datasets acquired over the cities of Trento and Civezzano with a ground resolution of two-centimeter demonstrated the effectiveness of the proposed model.

Download Full-text

k-Nearest Neighbor Learning with Graph Neural Networks

Mathematics ◽

10.3390/math9080830 ◽

2021 ◽

Vol 9 (8) ◽

pp. 830

Author(s):

Seokho Kang

Keyword(s):

Neural Network ◽

Nearest Neighbor ◽

Learning Algorithm ◽

Weighting Function ◽

High Sensitivity ◽

Training Data ◽

K Nearest Neighbor ◽

Main Challenge ◽

Benchmark Datasets ◽

Graph Neural Networks

k-nearest neighbor (kNN) is a widely used learning algorithm for supervised learning tasks. In practice, the main challenge when using kNN is its high sensitivity to its hyperparameter setting, including the number of nearest neighbors k, the distance function, and the weighting function. To improve the robustness to hyperparameters, this study presents a novel kNN learning method based on a graph neural network, named kNNGNN. Given training data, the method learns a task-specific kNN rule in an end-to-end fashion by means of a graph neural network that takes the kNN graph of an instance to predict the label of the instance. The distance and weighting functions are implicitly embedded within the graph neural network. For a query instance, the prediction is obtained by performing a kNN search from the training data to create a kNN graph and passing it through the graph neural network. The effectiveness of the proposed method is demonstrated using various benchmark datasets for classification and regression tasks.

Download Full-text

Structured Gaussian Process Regression of Music Mood

Fundamenta Informaticae ◽

10.3233/fi-2020-1970 ◽

2020 ◽

Vol 176 (2) ◽

pp. 183-203

Author(s):

Santosh Chapaneri ◽

Deepak Jayaswal

Keyword(s):

Gaussian Process ◽

Computational Cost ◽

Gaussian Process Regression ◽

Gaussian Mixture ◽

Feature Representation ◽

Affective Content ◽

Variational Bayesian Inference ◽

Benchmark Datasets ◽

Valence And Arousal ◽

Structured Regression

Modeling the music mood has wide applications in music categorization, retrieval, and recommendation systems; however, it is challenging to computationally model the affective content of music due to its subjective nature. In this work, a structured regression framework is proposed to model the valence and arousal mood dimensions of music using a single regression model at a linear computational cost. To tackle the subjectivity phenomena, a confidence-interval based estimated consensus is computed by modeling the behavior of various annotators (e.g. biased, adversarial) and is shown to perform better than using the average annotation values. For a compact feature representation of music clips, variational Bayesian inference is used to learn the Gaussian mixture model representation of acoustic features and chord-related features are used to improve the valence estimation by probing the chord progressions between chroma frames. The dimensionality of features is further reduced using an adaptive version of kernel PCA. Using an efficient implementation of twin Gaussian process for structured regression, the proposed work achieves a significant improvement in R2 for arousal and valence dimensions relative to state-of-the-art techniques on two benchmark datasets for music mood estimation.

Download Full-text

A Lightweight Dense Connected Approach with Attention on Single Image Super-Resolution

Electronics ◽

10.3390/electronics10111234 ◽

2021 ◽

Vol 10 (11) ◽

pp. 1234

Author(s):

Lei Zha ◽

Yu Yang ◽

Zicheng Lai ◽

Ziwei Zhang ◽

Juan Wen

Keyword(s):

Receptive Fields ◽

Super Resolution ◽

Feature Representation ◽

Single Image ◽

Deep Model ◽

Benchmark Datasets ◽

Rich Information ◽

Image Super Resolution ◽

Model Training ◽

Single Image Super Resolution

In recent years, neural networks for single image super-resolution (SISR) have applied more profound and deeper network structures to extract extra image details, which brings difficulties in model training. To deal with deep model training problems, researchers utilize dense skip connections to promote the model’s feature representation ability by reusing deep features of different receptive fields. Benefiting from the dense connection block, SRDensenet has achieved excellent performance in SISR. Despite the fact that the dense connected structure can provide rich information, it will also introduce redundant and useless information. To tackle this problem, in this paper, we propose a Lightweight Dense Connected Approach with Attention for Single Image Super-Resolution (LDCASR), which employs the attention mechanism to extract useful information in channel dimension. Particularly, we propose the recursive dense group (RDG), consisting of Dense Attention Blocks (DABs), which can obtain more significant representations by extracting deep features with the aid of both dense connections and the attention module, making our whole network attach importance to learning more advanced feature information. Additionally, we introduce the group convolution in DABs, which can reduce the number of parameters to 0.6 M. Extensive experiments on benchmark datasets demonstrate the superiority of our proposed method over five chosen SISR methods.

Download Full-text