scholarly journals Data Augmentation Algorithm Based on Generative Antagonism Networks (GAN) Model for Optical Transmission Networks (OTN)

2021 ◽  
Author(s):  
Liang Chen ◽  
Kunpeng Zheng ◽  
Yang Li ◽  
Xuelian Yang ◽  
Han Zhang ◽  
...  

OTN (Optical Transmission Networks) is one of the mainstream infrastructures over the ground-transmission networks, with the characteristics of large bandwidth, low delay, and high reliability. To ensure a stable working of OTN, it is necessary to preform high-level accurate functions of data traffic analysis, alarm prediction, and fault location. However, there is a serious problem for the implementation of these functions, caused by the shortage of available data but a rather-large amount of dirty data existed in OTN. In the view of current pretreatment, the extracted amount of effective data is very less, not enough to support machine learning. To solve this problem, this paper proposes a data augmentation algorithm based on deep learning. Specifically, Data Augmentation for Optical Transmission Networks under Multi-condition constraint (MVOTNDA) is designed based on GAN Mode with the demonstration of variable-length data augmentation method. Experimental results show that MVOTNDA has better performances than the traditional data augmentation algorithms.

2020 ◽  
Vol 10 (3) ◽  
pp. 62
Author(s):  
Tittaya Mairittha ◽  
Nattaya Mairittha ◽  
Sozo Inoue

The integration of digital voice assistants in nursing residences is becoming increasingly important to facilitate nursing productivity with documentation. A key idea behind this system is training natural language understanding (NLU) modules that enable the machine to classify the purpose of the user utterance (intent) and extract pieces of valuable information present in the utterance (entity). One of the main obstacles when creating robust NLU is the lack of sufficient labeled data, which generally relies on human labeling. This process is cost-intensive and time-consuming, particularly in the high-level nursing care domain, which requires abstract knowledge. In this paper, we propose an automatic dialogue labeling framework of NLU tasks, specifically for nursing record systems. First, we apply data augmentation techniques to create a collection of variant sample utterances. The individual evaluation result strongly shows a stratification rate, with regard to both fluency and accuracy in utterances. We also investigate the possibility of applying deep generative models for our augmented dataset. The preliminary character-based model based on long short-term memory (LSTM) obtains an accuracy of 90% and generates various reasonable texts with BLEU scores of 0.76. Secondly, we introduce an idea for intent and entity labeling by using feature embeddings and semantic similarity-based clustering. We also empirically evaluate different embedding methods for learning good representations that are most suitable to use with our data and clustering tasks. Experimental results show that fastText embeddings produce strong performances both for intent labeling and on entity labeling, which achieves an accuracy level of 0.79 and 0.78 f1-scores and 0.67 and 0.61 silhouette scores, respectively.


2015 ◽  
Vol 7 (12) ◽  
pp. 1126 ◽  
Author(s):  
Houman Rastegarfar ◽  
Daniel C. Kilper ◽  
Madeleine Glick ◽  
Nasser Peyghambarian

Symmetry ◽  
2019 ◽  
Vol 12 (1) ◽  
pp. 28 ◽  
Author(s):  
Chao Wang

In order to improve the accuracy of semantic model intrinsic detection, a skeleton-based high-level semantic model intrinsic self-symmetry detection method is proposed. The semantic analysis of the model set is realized by the uniform segmentation of the model within the same style, the component correspondence of the model between different styles, and the shape content clustering. Based on the results of clustering analysis, for a given three-dimensional (3D) point cloud model, according to the curve skeleton, the skeleton point pairs reflecting the symmetry between the model surface points are obtained by the election method, and the symmetry is extended to the model surface vertices according to these skeleton point pairs. With the help of skeleton, the symmetry of the point cloud model is obtained, and then the symmetry region of point cloud model is obtained by the symmetric correspondence matrix and spectrum method, so as to realize the intrinsic symmetry detection of the model. The experimental results show that the proposed method has the advantages of less time, high accuracy, and high reliability.


1981 ◽  
Vol 5 ◽  
pp. 5-8
Author(s):  
Pamela A. Geisler ◽  
J. France

At the present time, with the high reliability and performance of computer hardware, computer systems applied in any field must be judged more by the quality of the software provided. Thus it is highly relevant in an investigation of the use of computers in a field such as animal production, to concentrate on aspects of the software.Software provides the computer with the ability to obey instructions and to do as the user wishes. However, before arriving at these ‘machine instructions’ a number of steps have to be covered. First, it is essential to design the software — that is, to establish the requirements to be achieved on the computer. This design stage is followed by the implementation phase, in which the requirements as stated in English are transformed into such instructions as the machine can read and obey. The final phase is testing, in which it must be determined whether the requirements have been met, and to modify the design and iterate until the performance is satisfactory.Software in general can be divided into three classes — systems, utility and applications software. The systems software drives the machine and its associated peripherals such as a VDU and printer. The systems software also includes a file system for organization of the data on the relevant storage media (floppy disks, cartridges, magnetic tape). Also considered part of the systems software are the assembler, interpreters and compilers for high level languages such as BASIC and FORTRAN and for programming aids such as DEBUGGERS. Systems software is normally supplied with the computer and needs to be evaluated along with the hardware by any prospective purchaser of a computer system.


1997 ◽  
Vol 08 (01) ◽  
pp. 113-126 ◽  
Author(s):  
Sorin Draghici

This paper presents a neural network based artificial vision system able to analyze the image of a car given by a camera, locate the registration plate and recognize the registration number of the car. The paper describes in detail various practical problems encountered in implementing this particular application and the solution used to solve them. The main features of the system presented are: controlled stability-plasticity behavior, controlled reliability threshold, both off-line and on-line learning, self assessment of the output reliability and high reliability based on high level multiple feedback. The system has been designed using a modular approach. Sub-modules can be upgraded and/or substituted independently, thus making the system potentially suitable in a large variety of vision applications. The OCR engine was designed as an interchangeable plug-in module. This allows the user to choose an OCR engine which is suited to the particular application and to upgrade it easily in the future. At present, there are several versions of this OCR engine. One of them is based on a fully connected feedforward artificial neural network with sigmoidal activation functions. This network can be trained with various training algorithms such as error backpropagation. An alternative OCR engine is based on the constraint based decomposition (CBD) training architecture. The system has showed the following performances (on average) on real-world data: successful plate location and segmentation about 99%, successful character recognition about 98% and successful recognition of complete registration plates about 80%


Author(s):  
Maram Saudy ◽  
Safwan Khedr

Asphalt plays a significant role in pavement quality. The need for high-performance pavements with long service life and low maintenance requirements is the motive behind thorough research and studies of asphalt characteristics. This research focuses on studying all sources of Egyptian asphalt over a span of time using both conventional and Superpave grading techniques in order to characterize asphalt performance and also to answer the question whether the Egyptian asphalts need modification. The results of this research indicate that all Egyptian normal (virgin) 60/70 asphalt samples from different sources failed to meet penetration grading requirements, viscosity grading standards AC-20 (high quality); with minor exceptions, viscosity grading system AC-20 (low quality), and both AR-8000 and AR-1000 Aged Residue grading systems. When Superpave grading system was employed, results indicate that all normal asphalt samples failed to meet the basic requirements (without traffic adjustment) according to the Egyptian climatic requirements for high reliability projects (PG70-10 and PG76-10). The testing results accommodate Superpave requirements for lower levels of reliability and/or lower level of conservativeness. This emphasizes the flexibility and reliability of Superpave grading system as compared to conventional grading systems. On the other hand all modified asphalt samples, using an SBS modifier, passed according to the base high reliability projects and/or high level of conservativeness requirements of the Superpave grading system. Finally it is concluded that Egyptian asphalt should be modified in order to provide satisfactory performance especially for high reliability projects in hot regions with high and/or slow traffic.


2021 ◽  
Author(s):  
Binghua Li ◽  
Zhiwen Zhang ◽  
Feng Duan ◽  
Zhenglu Yang ◽  
Qibin Zhao ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document