scholarly journals A data-driven neural network architecture for sentiment analysis

2019 ◽  
Vol 53 (1) ◽  
pp. 2-19 ◽  
Author(s):  
Erion Çano ◽  
Maurizio Morisio
Keyword(s):  
Neural Network ◽  
Sentiment Analysis ◽  
Network Architecture ◽  
Network Models ◽  
Data Sets ◽  
Feature Maps ◽  
Neural Network Architecture ◽  
Neural Network Models ◽  
Content Type ◽  
Max Pooling

Purpose The fabulous results of convolution neural networks in image-related tasks attracted attention of text mining, sentiment analysis and other text analysis researchers. It is, however, difficult to find enough data for feeding such networks, optimize their parameters, and make the right design choices when constructing network architectures. The purpose of this paper is to present the creation steps of two big data sets of song emotions. The authors also explore usage of convolution and max-pooling neural layers on song lyrics, product and movie review text data sets. Three variants of a simple and flexible neural network architecture are also compared. Design/methodology/approach The intention was to spot any important patterns that can serve as guidelines for parameter optimization of similar models. The authors also wanted to identify architecture design choices which lead to high performing sentiment analysis models. To this end, the authors conducted a series of experiments with neural architectures of various configurations. Findings The results indicate that parallel convolutions of filter lengths up to 3 are usually enough for capturing relevant text features. Also, max-pooling region size should be adapted to the length of text documents for producing the best feature maps. Originality/value Top results the authors got are obtained with feature maps of lengths 6–18. An improvement on future neural network models for sentiment analysis could be generating sentiment polarity prediction of documents using aggregation of predictions on smaller excerpt of the entire text.

Spatial Variability Aware Deep Neural Networks (SVANN): A General Approach

2021 ◽  
Vol 12 (6) ◽  
pp. 1-21
Author(s):  
Jayant Gupta ◽  
Carl Molnar ◽  
Yiqun Xie ◽  
Joe Knight ◽  
Shashi Shekhar
Keyword(s):  
Neural Network ◽  
Spatial Variability ◽  
Network Architecture ◽  
Network Models ◽  
Neural Network Architecture ◽  
Neural Network Models ◽  
Climatic Zones ◽  
The Neural Network ◽  
Plant Hardiness ◽  
Interpretation Model

Spatial variability is a prominent feature of various geographic phenomena such as climatic zones, USDA plant hardiness zones, and terrestrial habitat types (e.g., forest, grasslands, wetlands, and deserts). However, current deep learning methods follow a spatial-one-size-fits-all (OSFA) approach to train single deep neural network models that do not account for spatial variability. Quantification of spatial variability can be challenging due to the influence of many geophysical factors. In preliminary work, we proposed a spatial variability aware neural network (SVANN-I, formerly called SVANN ) approach where weights are a function of location but the neural network architecture is location independent. In this work, we explore a more flexible SVANN-E approach where neural network architecture varies across geographic locations. In addition, we provide a taxonomy of SVANN types and a physics inspired interpretation model. Experiments with aerial imagery based wetland mapping show that SVANN-I outperforms OSFA and SVANN-E performs the best of all.

scholarly journals Developing a seq2seq neural network using visual attention to transform mathematical expressions from images to LaTeX.

Doklady BGUIR ◽  
2022 ◽  
Vol 19 (8) ◽  
pp. 40-44
Author(s):  
P. A. Vyaznikov ◽  
I. D. Kotilevets
Keyword(s):  
Neural Network ◽  
Visual Attention ◽  
Network Architecture ◽  
Network Models ◽  
Neural Network Architecture ◽  
Neural Network Models ◽  
The Neural Network ◽  
Mathematical Expressions ◽  
Series Of Experiments ◽  
Visual Attention Mechanism

The paper presents the methods of development and the results of research on the effectiveness of the seq2seq neural network architecture using Visual Attention mechanism to solve the im2latex problem. The essence of the task is to create a neural network capable of converting an image with mathematical expressions into a similar expression in the LaTeX markup language. This problem belongs to the Image Captioning type: the neural network scans the image and, based on the extracted features, generates a description in natural language. The proposed solution uses the seq2seq architecture, which contains the Encoder and Decoder mechanisms, as well as Bahdanau Attention. A series of experiments was conducted on training and measuring the effectiveness of several neural network models.

scholarly journals Novel convolutional neural networks for efficient classification of rotated and scaled images

2021 ◽  
Author(s):  
Paweł Tarasiuk ◽  
Piotr S. Szczepaniak
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Network Architecture ◽  
Data Augmentation ◽  
Network Models ◽  
Input Image ◽  
Feature Maps ◽  
Neural Network Models ◽  
Geometric Transformations

AbstractThis paper presents a novel method for improving the invariance of convolutional neural networks (CNNs) to selected geometric transformations in order to obtain more efficient image classifiers. A common strategy employed to achieve this aim is to train the network using data augmentation. Such a method alone, however, increases the complexity of the neural network model, as any change in the rotation or size of the input image results in the activation of different CNN feature maps. This problem can be resolved by the proposed novel convolutional neural network models with geometric transformations embedded into the network architecture. The evaluation of the proposed CNN model is performed on the image classification task with the use of diverse representative data sets. The CNN models with embedded geometric transformations are compared to those without the transformations, using different data augmentation setups. As the compared approaches use the same amount of memory to store the parameters, the improved classification score means that the proposed architecture is more optimal.

Artificial neural network and regression models for prediction of sewing thread consumption for multilayered fabric assembly at lockstitch 301 seam

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Md Vaseem Chavhan ◽  
M. Ramesh Naidu ◽  
Hayavadana Jamakhandi
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Regression Model ◽  
Regression Models ◽  
Network Models ◽  
Quadratic Regression ◽  
Neural Network Models ◽  
Content Type ◽  
Sewing Thread ◽  
Artificial Neural

Purpose This paper aims to propose the artificial neural network (ANN) and regression models for the estimation of the thread consumption at multilayered seam assembly stitched with lock stitch 301. Design/methodology/approach In the present study, the generalized regression and neural network models are developed by considering the fabric types: woven, nonwoven and multilayer combination thereof, with basic sewing parameters: sewing thread linear density, stitch density, needle count and fabric assembly thickness. The network with feed-forward backpropagation is considered to build the ANN, and the training function trainlm of MATLAB software is used to adjust weight and basic values according to the optimization of Levenberg Marquardt. The performance of networks measured in terms of the mean squared error and the layer output is set according to the sigmoid transfer function. Findings The proposed ANN and regression model are able to predict the thread consumption with more accuracy for multilayered seam assembly. The predictability of thread consumption from available geometrical models, regression models and industrial empirical techniques are compared with proposed linear regression, quadratic regression and neural network models. The proposed quadratic regression model showed a good correlation with practical thread consumption value and more accuracy in prediction with an overall 4.3% error, as compared to other techniques for given multilayer substrates. Further, the developed ANN network showed good accuracy in the prediction of thread consumption. Originality/value The estimation of thread consumed while stitching is the prerequisite of the garment industry for inventory management especially with the introduction of the costly high-performance sewing thread. In practice, different types of fabrics are stitched at multilayer combinations at different locations of the stitched product. The ANN and regression models are developed for multilayered seam assembly of woven and nonwoven fabric blend composition for better prediction of thread consumption.

Neural network models for actual duration of Greek highway projects

2019 ◽  
Vol 17 (6) ◽  
pp. 1323-1339
Author(s):  
Magdalini Titirla ◽  
Georgios Aretoulis
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Correlation Analysis ◽  
Network Models ◽  
Neural Network Models ◽  
Tender Offer ◽  
Highway Projects ◽  
Content Type ◽  
Initial Cost ◽  
Actual Duration

Purpose This paper aims to examine selected similar Greek highway projects to create artificial neural network-based models to predict their actual construction duration based on data available at the bidding stage. Design/methodology/approach Relevant literature review is presented that highlights similar research approaches. Thirty-seven highway projects, constructed in Greece, with similar type of available data, were examined. Considering each project’s characteristics and the actual construction duration, correlation analysis is implemented, with the aid of SPSS. Correlation analysis identified the most significant project variables toward predicting actual duration. Furthermore, the WEKA application, through its attribute selection function, highlighted the most important subset of variables. The selected variables through correlation analysis and/or WEKA and appropriate combinations of these are used as input neurons for a neural network. Fast Artificial Neural Network (FANN) Tool is used to construct neural network models in an effort to predict projects’ actual duration. Findings Variables that significantly correlate with actual time at completion include initial cost, initial duration, length, lanes, technical projects, bridges, tunnels, geotechnical projects, embankment, landfill, land requirement (expropriation) and tender offer. Neural networks’ models succeeded in predicting actual completion time with significant accuracy. The optimum neural network model produced a mean squared error with a value of 6.96E-06 and was based on initial cost, initial duration, length, lanes, technical projects, tender offer, embankment, existence of bridges, geotechnical projects and landfills. Research limitations/implications The sample size is limited to 37 projects. These are extensive highway projects with similar work packages, constructed in Greece. Practical implications The proposed models could early in the planning stage predict the actual project duration. Originality/value The originality of the current study focuses both on the methodology applied (combination of Correlation Analysis, WEKA, FannTool) and on the resulting models and their potential application for future projects.

scholarly journals Data-to-Text Generation with Content Selection and Planning

2019 ◽  
Vol 33 ◽  
pp. 6908-6915 ◽  
Author(s):  
Ratish Puduppully ◽  
Li Dong ◽  
Mirella Lapata
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Large Scale ◽  
Network Models ◽  
Text Generation ◽  
Neural Network Models ◽  
Generation Task ◽  
Content Selection ◽  
End To End ◽  
Two Stages

Recent advances in data-to-text generation have led to the use of large-scale datasets and neural network models which are trained end-to-end, without explicitly modeling what to say and in what order. In this work, we present a neural network architecture which incorporates content selection and planning without sacrificing end-to-end training. We decompose the generation task into two stages. Given a corpus of data records (paired with descriptive documents), we first generate a content plan highlighting which information should be mentioned and in which order and then generate the document while taking the content plan into account. Automatic and human-based evaluation experiments show that our model1 outperforms strong baselines improving the state-of-the-art on the recently released RotoWIRE dataset.

scholarly journals Intelligent Prediction of Refrigerant Amounts Based on Internet of Things

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-12 ◽  
Author(s):  
Jincai Chang ◽  
Qiuling Pan ◽  
Zhihao Shen ◽  
Hao Qin
Keyword(s):  
Neural Network ◽  
Internet Of Things ◽  
Dimensionality Reduction ◽  
Rbf Neural Network ◽  
Network Models ◽  
Practical Significance ◽  
Data Sets ◽  
Effective Parameters ◽  
Neural Network Models ◽  
Refrigeration Unit

In a refrigeration unit, the amount of refrigerant has a substantial influence on the entire refrigeration system. To predict the amount of refrigerant in refrigerators with the best performance, this study used refrigerator data collected in real time via the Internet of Things, which were screened to include only the effective parameters related to the compressor and refrigeration properties (based on their practical significance and the research background) and cleaned by applying longitudinal dimensionality reduction and transverse dimensionality reduction. Then, on the basis of an idealized model for refrigerator data, a model of the relationships between refrigerant amount (the dependent variable) and temperature variation, refrigerator compartment temperature, freezer temperature, and other relevant parameters (independent variables) was established. A refrigeration model based on a neural network was then established for predicting the amount of refrigerant and was used to predict five unknown amounts of refrigerant from data sets. BP neural network and RBF neural network models were used to compare the prediction results and analyze the loss functions. From the results, it was concluded that the unknown amount of refrigerant was most likely to be 32.5 g. It is of great practical significance for refrigerator production and maintenance to study the prediction of the amount of refrigerant remaining in a refrigerator.

scholarly journals ORTHOSEG: A DEEP MULTIMODAL CONVOLUTONAL NEURAL NETWORK ARCHITECTURE FOR SEMANTIC SEGMENTATION OF ORTHOIMAGERY

2018 ◽  
Vol XLII-5 ◽  
pp. 621-628 ◽  
Author(s):  
P. Bodani ◽  
K. Shreshtha ◽  
S. Sharma
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Multiple Scales ◽  
Semantic Segmentation ◽  
Effective Field ◽  
Surface Model ◽  
Training Procedure ◽  
Feature Maps ◽  
Neural Network Architecture ◽  
Wide Range

<p><strong>Abstract.</strong> This paper addresses the task of semantic segmentation of orthoimagery using multimodal data e.g. optical RGB, infrared and digital surface model. We propose a deep convolutional neural network architecture termed OrthoSeg for semantic segmentation using multimodal, orthorectified and coregistered data. We also propose a training procedure for supervised training of OrthoSeg. The training procedure complements the inherent architectural characteristics of OrthoSeg for preventing complex co-adaptations of learned features, which may arise due to probable high dimensionality and spatial correlation in multimodal and/or multispectral coregistered data. OrthoSeg consists of parallel encoding networks for independent encoding of multimodal feature maps and a decoder designed for efficiently fusing independently encoded multimodal feature maps. A softmax layer at the end of the network uses the features generated by the decoder for pixel-wise classification. The decoder fuses feature maps from the parallel encoders locally as well as contextually at multiple scales to generate per-pixel feature maps for final pixel-wise classification resulting in segmented output. We experimentally show the merits of OrthoSeg by demonstrating state-of-the-art accuracy on the ISPRS Potsdam 2D Semantic Segmentation dataset. Adaptability is one of the key motivations behind OrthoSeg so that it serves as a useful architectural option for a wide range of problems involving the task of semantic segmentation of coregistered multimodal and/or multispectral imagery. Hence, OrthoSeg is designed to enable independent scaling of parallel encoder networks and decoder network to better match application requirements, such as the number of input channels, the effective field-of-view, and model capacity.</p>

Sign in / Sign up

Export Citation Format

Share Document