Transformer architectures are highly expressive because they use self-attention mechanisms to encode long-range dependencies in the input sequences. In this paper, we present a literature review on Transformer-based (TB) models, providing a detailed overview of each model in comparison to the Transformer’s standard architecture. This survey focuses on TB models used in the field of Natural Language Processing (NLP) for textual-based tasks. We begin with an overview of the fundamental concepts at the heart of the success of these models. Then, we classify them based on their architecture and training mode. We compare the advantages and disadvantages of popular techniques in terms of architectural design and experimental value. Finally, we discuss open research, directions, and potential future work to help solve current TB application challenges in NLP.

/pdf/end-to-end-transformer-based-models-in-textual-based-nlp-3i48a8ug.pdf

End-to-End Transformer-Based Models in Textual-Based NLP

The hotel uses modern information technology to establish a customer data group, which contains information about the customer's information on space configuration, color design, etc., and uses scientific operation methods to accelerate the rapid, transformation and upgrading of the hotel. The rapid development of information technology represented by the big intelligent mobile cloud has a lot of functions, can meet the needs of different customers and diversified characteristics at the same time, and provides a complete, stable and accurate information integration platform for hotel system management. The main research directions of DASH include client adaptive algorithm design, multi-client fairness and resource utilization issues, client improvement under multiple servers, and server load balancing. The development and application of scientific and technological information has broken through the development path of the hotel management system, providing new development space and opportunities for the improvement of the hotel management system, but at the same time it will also face more new challenges. Therefore, this article focuses on the characteristics of the tourism accommodation management system under the big data environment, and analyzes the development of the hotel management system under the network environment. The purpose is to provide some reference and reference significance.

Application of DASH client optimization and artificial intelligence in the management and operation of big data tourism hotels

The hotel uses modern information technology to establish a customer data group, which contains information about the customer's information on space configuration, color design, etc., and uses scientific operation methods to accelerate the rapid, transformation and upgrading of the hotel. The rapid development of information technology represented by the big intelligent mobile cloud has a lot of functions, can meet the needs of different customers and diversified characteristics at the same time, and provides a complete, stable and accurate information integration platform for hotel system management. The main research directions of DASH include client adaptive algorithm design, multi-client fairness and resource utilization issues, client improvement under multiple servers, and server load balancing. The development and application of scientific and technological information has broken through the development path of the hotel management system, providing new development space and opportunities for the improvement of the hotel management system, but at the same time it will also face more new challenges. Therefore, this article focuses on the characteristics of the tourism accommodation management system under the big data environment, and analyzes the development of the hotel management system under the network environment. The purpose is to provide some reference and reference significance. 

Due to the low compression performance of traditional compression models, we have developed a new HOA based Fractional KCA with 2D-DWT for improving the multispectral image quality In this paper, we present a novel multispectral image compression method for improving the complexity by maintaining quality reconstruction and also reducing the size of the storage of multispectral images Initially, Karhunen–Loeve transform (KLT) is used to remove the spatial redundancies In the second stage, 2D DWT is used to eliminate the intraband spatial redundancies In the third stage, Fractional KCA (FKCA) is applied to improve the post-transformation process FKCA is connected to the band of all wavelet sub-bands to minimize the spatial redundancy between intra sub-bands Finally, the Hybrid Optimal algorithm (HOA) based FKCA is used to eliminate the residual and information redundancy among the neighboring bands The experimental analysis of proposed 2D-DWT based Fractional KCA shows that the model improves the performance of compression data in terms of PSNR, MSSI, and VIF Also, the multispectral image dataset shows the proposed compression model outperforms the existing compression models such as FKLT + PCA, ADWT + OADL, and DWT + DCT

Hybrid optimal algorithm-based 2D discrete wavelet transform for image compression using fractional KCA

Digital communication is the physical transfer of data through a point-to-point or point-to-point multidisciplinary communication channel. This is to exchange private messages. Digital communication plays an important role in today's world of electronics. The rate of data transfer in digital communication depends on its characteristics Digital communication provides a seamless experience to customers and partners direct communication and AI chat bots and automation Digital in various forms such as digital makes communication easier for customers to access companies simultaneously. It's convenient, it's easy, cheap, and fast because it can be done over long distances over the Internet and other things can be done via digital hardware processing circuits. Physical transmission of data through point-to-point multidisciplinary communication channels is more flexible than analog with higher energy consumption. It required more bandwidth compared to analog systems.

Exploring Various Digital Communication and its Classification

Segmentation and classification of breast cancer using novel deep learning architecture

Content-based image retrieval has become popular in the retrieval of images from large image database using reduced human intervention. Researchers are still in need to develop effective systems for dealing many of the wide-scope scientific and medical applications. Past research works have faced a problem on differentiating different images by means of using the single features alone. In this paper, a multi-level matching scheme is introduced for retrieval of image based on a hybrid feature similarity integrating local and global features. Both global- and local-level features included in multi-level scheme are used for image representation. From an image, the color information is extracted globally using a new color, edge directivity descriptor and color-based features. Further, the interest of points from each image is detected using local descriptors called local binary pattern and speeded-up robust features. Using two image databases, the improved retrieval accuracy obtained with the combination of global and local features is analyzed. Experimental outcomes have revealed the effectiveness of proposed system on achieving 91% and 92% precision rates over two datasets compared to other existing methods.

Efficient hybrid multi-level matching with diverse set of features for image retrieval

Transfer learning based recurrent neural network algorithm for linguistic analysis

In image applications, noise removal is one of the important requirement. Impulse noise is removed by median filters using image processing. Median filter is the good non-linear filter used for edge preservation in images and in real-time hardware implementation, sorting network is used. Sorting network must have an efficient implementation in terms of area, delay and power. Row three cell sorter, column three cell sorter, and right diagonal three cell sorter comprise the median filter. The median value of the inputs is the output of the median filter. Three data comparators give the maximum, minimum and middle value in the three cell sorter. The comparator returns the highest and lowest value of given inputs. The middle value is calculated using a modified shear sorting method. This paper compares two existing data comparators and suggests a modified version of those two existing comparators in factors of delay and area. Full adders and full subtractors are used to build the proposed comparators using MUX. The two’s complement based data comparator consumes less time. As a result, an implementation using two’s complement is suited for high-speed implementations. The carry select data comparator based implementation consumes smaller area, hence it can be employed in reduced area applications. The median filter is intended to be used with a two’s complement based data comparator. In the parallel architecture of modified shear sorting for constructing median filter, different architectures for data comparators were used.

V. Anbumani

Papers

Segmentation and classification of breast cancer using novel deep learning architecture

Efficient hybrid multi-level matching with diverse set of features for image retrieval

Transfer learning based recurrent neural network algorithm for linguistic analysis

Hybrid optimal algorithm-based 2D discrete wavelet transform for image compression using fractional KCA

An Efficient VLSI design of Median Filters using 8-bit Data Comparators in Image Applications