Journal on Big Data

About the Journal

Journal on Big Data is launched in a new area when the engineering features of big data are setting off upsurges of explorations in algorithms, raising challenges on big data, and industrial development integration; and novel paradigms in this cross –disciplinary field need to be constructed by translating complex innovative ideas from various fields.

  • l1-norm Based GWLP for Robust Frequency Estimation
  • Abstract In this work, we address the frequency estimation problem of a complex singletone embedded in the heavy-tailed noise. With the use of the linear prediction (LP) property and l1-norm minimization, a robust frequency estimator is developed. Since the proposed method employs the weighted l1-norm on the LP errors, it can be regarded as an extension of the lp-generalized weighted linear predictor. Computer simulations are conducted in the environment of α-stable noise, indicating the superiority of the proposed algorithm, in terms of its robust to outliers and nearly optimal estimation performance. More
  •   Views:180       Downloads:150        Download PDF
  • The Analysis of China’s Integrity Situation Based on Big Data
  • Abstract In order to study deeply the prominent problems faced by China’s clean government work, and put forward effective coping strategies, this article analyzes the network information of anti-corruption related news events, which is based on big data technology. In this study, we take the news report from the website of the Communist Party of China (CPC) Central Commission for Discipline Inspection (CCDI) as the source of data. Firstly, the obtained text data is converted to word segmentation and stop words under preprocessing, and then the pre-processed data is improved by vectorization and text clustering, finally, after text clustering, the key… More
  •   Views:196       Downloads:146        Download PDF
  • On Visualization Analysis of Stock Data
  • Abstract Big data technology is changing with each passing day, generating massive amounts of data every day. These data have large capacity, many types, fast growth, and valuable features. The same is true for the stock investment market. The growth of the amount of stock data generated every day is difficult to predict. The price trend in the stock market is uncertain, and the valuable information hidden in the stock data is difficult to detect. For example, the price trend of stocks, profit trends, how to make a reasonable speculation on the price trend of stocks and profit trends is a… More
  •   Views:168       Downloads:173        Download PDF
  • A Privacy Preserving Deep Linear Regression Scheme Based on Homomorphic Encryption
  • Abstract This paper proposes a strategy for machine learning in the ciphertext domain. The data to be trained in the linear regression equation is encrypted by SHE homomorphic encryption, and then trained in the ciphertext domain. At the same time, it is guaranteed that the error of the training results between the ciphertext domain and the plaintext domain is in a controllable range. After the training, the ciphertext can be decrypted and restored to the original plaintext training data. More
  •   Views:191       Downloads:134        Download PDF
  • A Meaningful Image Encryption Algorithm Based on Prediction Error and Wavelet Transform
  • Abstract Image encryption (IE) is a very useful and popular technology to protect the privacy of users. Most algorithms usually encrypt the original image into an image similar to texture or noise, but texture and noise are an obvious visual indication that the image has been encrypted, which is more likely to cause the attacks of enemy. To overcome this shortcoming, many image encryption systems, which convert the original image into a carrier image with visual significance have been proposed. However, the generated cryptographic image still has texture features. In line with the idea of improving the visual quality of the… More
  •   Views:228       Downloads:177        Download PDF