Journal of Visual Communication and Image Representation

This study analyzed 928 full-text research articles retrieved from DESIDOC Journal of Library and Information Technology for the period of 1981–2018 using Latent Dirichlet Allocation. The study further tagged the articles with the modeled topics. 50 core topics were identified throughout the period of 38 years whereas only 26 topics were unique in nature. Bibliometrics, ICT, information retrieval, and user studies were highly researched areas in India for the epoch. Further, Spain and Taiwan showed common research trends and areas as India whereas India has quite distinct research interests from America and China. Therefore, researchers in Library and Information Science in India should pay more attention to the topics which are under-researched. Further, it was found that there were some unique sub-fields to Indian Library and Information Science research, such as open access; online exhibition; virtual libraries; multimedia libraries; open source software; library automation; and library management system. With the passage of time topics evolve over time, new topics emerge, and old ones become obsolete. Topic modeling not only helps the researcher to determine the trending themes or related fields with respect to their field of interest but also helps them to identify new concepts and fields over time.

Mapping of topics in DESIDOC Journal of Library and Information Technology, India: a study

Correction to Annals of Applied Statistics 1 (2007) 17--35 [doi:10.1214/07-AOAS114]

Correction: A correlated topic model of Science

Research Advances in the Integration of Big Data and Smart Computing

Medical image data, like most patient information, have a strong requirement for privacy and confidentiality. This makes transmitting medical image data, within an open network, problematic, due to the aforementioned issues, along with the dangers of data/information leakage. Possible solutions in the past have included the utilization of information-hiding and image-encryption technologies; however, these methods can cause difficulties when attempting to recover the original images. In this work, we developed an algorithm for protecting medical image key regions. Coefficient of variation is first employed to identify key regions, a.k.a. image lesion areas; then additional areas are processed as blocks and texture complexity is analyzed. Next, our novel reversible data-hiding algorithm embeds lesion area contents into a high-texture area, after which an Arnold transformation is utilized to protect the original lesion information. After this, we use image basic information ciphertext and decryption parameters to generate a quick response (QR) code used in place of original key regions. The approach presented here allows for the storage (and sending) of medical image data within open network environments, while ensuring only authorized personnel are able to recover sensitive patient information (both image and meta-data) without information loss. Peak signal to noise ratio and the Structural Similarity Index measures show that the algorithm presented in this work can encrypt and restore original images without information loss. Moreover, by adjusting the threshold and the Mean Squared Error, we can control the overall quality of the image: the higher the threshold, the better the quality and vice versa. This allows the encryptor to control the amount of degradation as, at appropriate amounts, degradation aids in the protection of the image. As shown in the experimental results, the proposed method allows for (a) the safe transmission and storage of medical image data, (b) the full recovery (no information loss) of sensitive regions within the medical image following encryption, and (c) meta-data about the patient and image to be stored within and recovered from the public image.

/pdf/a-partial-encryption-algorithm-for-medical-images-based-on-2xlgod80g5.pdf

A partial encryption algorithm for medical images based on quick response code and reversible data hiding technology

With the explosive growth of academic writing, it is difficult for researchers to find significant papers in their area of interest. In this paper, we propose a pipeline model, named collective topical PageRank, to evaluate the topic-dependent impact of scientific papers. First, we fit the model to a correlation topic model based on the textual content of papers to extract scientific topics and correlations. Then, we present a modified PageRank algorithm, which incorporates the venue, the correlations of the scientific topics, and the publication year of each paper into a random walk to evaluate the paper’s topic-dependent academic impact. Our experiments showed that the model can effectively identify significant papers as well as venues for each scientific topic, recommend papers for further reading or citing, explore the evolution of scientific topics, and calculate the venues’ dynamic topic-dependent academic impact.

Collective topical PageRank: a model to evaluate the topic-dependent academic impact of scientific papers

This﻿article﻿describes﻿how﻿text﻿documents﻿are﻿a﻿major﻿data﻿structure﻿in﻿the﻿era﻿of﻿big﻿data.﻿With﻿the﻿ explosive﻿growth﻿of﻿data,﻿the﻿number﻿of﻿documents﻿with﻿multi-labels﻿has﻿increased﻿dramatically.﻿ The﻿popular﻿multi-label﻿classification﻿technology,﻿which﻿is﻿usually﻿employed﻿to﻿handle﻿multinomial﻿ text﻿documents,﻿is﻿sensitive﻿to﻿the﻿noise﻿terms﻿of﻿text﻿documents.﻿Therefore,﻿there﻿still﻿exists﻿a﻿huge﻿ room﻿ for﻿ multi-label﻿ classification﻿ of﻿ text﻿ documents.﻿ This﻿ article﻿ introduces﻿ a﻿ supervised﻿ topic﻿ model,﻿named﻿labeled﻿LDA﻿with﻿function﻿terms﻿(LF-LDA),﻿to﻿filter﻿out﻿the﻿noisy﻿function﻿terms﻿ from﻿text﻿documents,﻿which﻿can﻿help﻿to﻿improve﻿the﻿performance﻿of﻿multi-label﻿classification﻿of﻿text﻿ documents.﻿The﻿article﻿also﻿shows﻿the﻿derivation﻿of﻿the﻿Gibbs﻿Sampling﻿formulas﻿in﻿detail,﻿which﻿ can﻿be﻿generalized﻿to﻿other﻿similar﻿topic﻿models.﻿Based﻿on﻿the﻿textual﻿data﻿set﻿RCV1-v2,﻿the﻿article﻿ compared﻿the﻿proposed﻿model﻿with﻿other﻿two﻿state-of-the-art﻿multi-label﻿classifiers,﻿Tuned﻿SVM﻿and﻿ labeled﻿LDA,﻿on﻿both﻿Macro-F1﻿and﻿Micro-F1﻿metrics.﻿The﻿result﻿shows﻿that﻿LF-LDA﻿outperforms﻿ them﻿and﻿has﻿the﻿lowest﻿variance,﻿which﻿indicates﻿the﻿robustness﻿of﻿the﻿LF-LDA﻿classifier. KeyWoRDS Function Terms, Gibbs Sampling, Graph Model, Multi Label, Parameter Estimation, Probability Generation Process, Text Classification, Topic Model

LF-LDA: A Supervised Topic Model for Multi-Label Documents Classification

The privacy issue in health care application is more important. This paper presents an application in medical image for protecting patient information (called as a medical record). The reversible image-hiding scheme based on histogram shifting is mainly used for protecting medical record. In addition, we applied two parameters, linear predictor with non-uniform weight and threshold, to improve the performance in terms of medical image quality and embedding record ratio. The weight is adapted to improve the prediction accuracy, thus increasing confidential information embedded capacity. A threshold is used to determine image characteristics and avoid overflow or underflow problems. For image characteristic decisions, we choose a peak point and then, shifted right one unit for data concealing. Before the data hiding, our method will be also pre-process the underflow and overflow problems for the image in which the embedded medical record overflow occurred in the image and for the image generation has no salt and pepper effect. Experimental results show that our method is capable of providing a great embedding capacity without making noticeable distortion. In our method, using weight and threshold values can indeed increase the embedding capacity from histogram-based data hiding while retaining the image quality.

Yongtao Yu

Papers

Collective topical PageRank: a model to evaluate the topic-dependent academic impact of scientific papers

LF-LDA: A Supervised Topic Model for Multi-Label Documents Classification

Reversible Data Hiding Based on Medical Image Systems by Means of Histogram Strategy