A Multiple-Input Neural Network Model for Predicting Cotton Production Quantity: A Case Study

Livieris, Ioannis E.; Dafnis, Spiros D.; Papadopoulos, George K.; Kalivas, Dionissios

doi:10.3390/a13110273

Cited by 15 publications

(8 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Particularly, we used the six PCs sets (computed by the GRM of each marker set) as six inputs feeding to the network in different input layers simultaneously. Other works have shown that a multiple input strategy can reduce overfitting and computational cost while at the same time exploits mixed data improving prediction ( Livieris et al 2020 , Xiong et al 2021 ). Thus, the network accepted six different input layers which independently forwards in six different hidden dense layers.…”

Section: Methodsmentioning

confidence: 99%

Evaluation of Deep Learning for predicting rice traits using structural and single-nucleotide genomic variants

Vourlaki,

Ramos-Onsins,

Pérez-Enciso

et al. 2024

Preprint

View full text Add to dashboard Cite

Structural variants (SVs) such as deletions, inversions, duplications, and Transposable El-ement (TE) Insertion Polymorphisms (TIPs) are prevalent in plant genomes and have played an important role in evolution and domestication, as they constitute a significant source of ge-nomic and phenotypic variability. Nevertheless, most methods in quantitative genetics that fo-cus on plant crop improvement, such as genomic prediction, consider Single Nucleotide Poly-morphisms (SNPs) as the only type of genetic marker. Here, we used rice to investigate wheth-er combining the structural and nucleotide genome-wide variation can improve prediction abil-ity of traits when compared to using only SNPs. Moreover, we also examine the potential ad-vantage of Deep Learning (DL) networks over Bayesian Linear models, which have been widely applied in genomic prediction. Specifically, the performance of BayesC and a Bayesian Repro-ducible Kernel Hilbert space regressions were compared to two different DL architectures, the Multilayer Perceptron, and the Convolution Neural Network. We further explore their prediction ability by using various marker input strategies and found that exploiting structural and nucleo-tide variation improves prediction ability on complex traits in rice. Also, DL models outper-formed Bayesian models in 75% of the studied cases. Finally, DL systematically improved pre-diction ability of binary traits against the Bayesian models.

show abstract

Section: Methodsmentioning

confidence: 99%

Evaluation of Deep Learning for predicting rice traits using structural and single-nucleotide genomic variants

Vourlaki,

Ramos-Onsins,

Pérez-Enciso

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…While SegNet [127] and Faster R-CNN [16] have been used for optical-MS image segmentation. On the other hand, RNN models (with LSTM or GRU) are usually selected when using temporal data [105], [106].…”

Section: What To Focus?mentioning

confidence: 99%

“…In this case, the comparison is categorical and corresponds to whether the fusion strategy shows the best, worst, or inbetween predictive performance. The main outcome is that feature-level fusion has better predictive performance compared with input and decision-level fusion in most of the cases [18], [32], [34]- [36], [58], [86], [101], [104], [105], [133], [200]. Furthermore, based on the analyzed evidence, the feature-level fusion performance stands above the worst performance alternatives.…”

Section: Bestmentioning

confidence: 99%

Common Practices and Taxonomy in Deep Multiview Fusion for Remote Sensing Applications

Mena,

Arenas,

Nuske

et al. 2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

The advances in remote sensing technologies have boosted applications for Earth observation. These technologies provide multiple observations or views with different levels of information. They might contain static or temporary views with different levels of resolution, in addition to having different types and amounts of noise due to sensor calibration or deterioration. A great variety of deep learning models have been applied to fuse the information from these multiple views, known as deep multiview or multi-modal fusion learning. However, the approaches in the literature vary greatly since different terminology is used to refer to similar concepts or different illustrations are given to similar techniques. This article gathers works on multi-view fusion for Earth observation by focusing on the common practices and approaches used in the literature. We summarize and structure insights from several different publications concentrating on unifying points and ideas. In this manuscript, we provide a harmonized terminology while at the same time mentioning the various alternative terms that are used in literature. The topics covered by the works reviewed focus on supervised learning with the use of neural network models. We hope this review, with a long list of recent references, can support future research and lead to a unified advance in the area.

show abstract

“…Notice that although a traditional deep neural network model is able to analyze and encode any complex function, the convergence of its training process may be degradated due to the number of weights, which exponentially increases as the number of layers increases and due to the vanishing gradient problem, which usually occurs in large networks. In contrast, the significant advantages of the proposed model's architecture is that it provides more flexibility and adaptivity for low computation effort compared to a fully connect neural network with a similar number of layers as well as greater resistance to the vanishing gradient problem, due to its sparse structure [16,20].…”

Section: Multiple-input Cryptocurrency Deep Learning Modelmentioning

confidence: 99%

“…The rationale for the utilization of a multi-input neural network is that these types of models have been originally proposed for more efficiently exploiting mixed data and refers to the case of having multiple types of independent data [14]. In the literature, these models have been successfully applied for addressing a variety of difficult real-world problems reporting promising results while they were found to outperform traditional single output models [14][15][16][17][18]. The main idea behind these models is to extract valuable information from each category of mixed data, independently and then concatenate the information for issuing the final prediction.…”

Section: Introductionmentioning

confidence: 99%

An Advanced CNN-LSTM Model for Cryptocurrency Forecasting

et al. 2021

Self Cite

View full text Add to dashboard Cite

Nowadays, cryptocurrencies are established and widely recognized as an alternative exchange currency method. They have infiltrated most financial transactions and as a result cryptocurrency trade is generally considered one of the most popular and promising types of profitable investments. Nevertheless, this constantly increasing financial market is characterized by significant volatility and strong price fluctuations over a short-time period therefore, the development of an accurate and reliable forecasting model is considered essential for portfolio management and optimization. In this research, we propose a multiple-input deep neural network model for the prediction of cryptocurrency price and movement. The proposed forecasting model utilizes as inputs different cryptocurrency data and handles them independently in order to exploit useful information from each cryptocurrency separately. An extensive empirical study was performed using three consecutive years of cryptocurrency data from three cryptocurrencies with the highest market capitalization i.e., Bitcoin (BTC), Etherium (ETH), and Ripple (XRP). The detailed experimental analysis revealed that the proposed model has the ability to efficiently exploit mixed cryptocurrency data, reduces overfitting and decreases the computational cost in comparison with traditional fully-connected deep neural networks.

show abstract

A Multiple-Input Neural Network Model for Predicting Cotton Production Quantity: A Case Study

Cited by 15 publications

References 31 publications

Evaluation of Deep Learning for predicting rice traits using structural and single-nucleotide genomic variants

Evaluation of Deep Learning for predicting rice traits using structural and single-nucleotide genomic variants

Common Practices and Taxonomy in Deep Multiview Fusion for Remote Sensing Applications

An Advanced CNN-LSTM Model for Cryptocurrency Forecasting

Contact Info

Product

Resources

About