Opinion Mining is a valuable knowledge resource to understand the collective opinions and to take better decisions. It is a Natural Language Processing (NLP) task that decides whether a text expresses positive or negative sentiment. Web contents are increasing rapidly and providing a huge number of information. It is an important research issue to analyze and organize these enormous information for better knowledge extraction. In this paper, we emphasis on opinion mining for Bangla text using web based diverse data. We apply both Linear and Nonlinear Support Vector Machine as machine learning technique and N -gram method to classify Bangla documents collected from social media sites. Most of works in this arena take a single word as a vector. Instead of thinking a single word as a vector, we used one vector containing more than one words using N-gram. N-grams of texts are extensively used in text mining and natural language processing tasks. We found better results using N-grams for different values of n.

N-Gram Based Sentiment Mining for Bangla Text Using Support Vector Machine

Detection and identification of microbial species are crucial in a wide range of industries, including production of beverages, foods, cosmetics, and pharmaceuticals. Traditionally, colony formation and its morphological analysis (e.g., size, shape, and color) with a naked eye have been employed for this purpose. However, such a conventional method is time consuming, labor intensive, and not very reproducible. To overcome these problems, we propose a novel method that detects microcolonies (diameter 10-500 μm) using a lensless imaging system. When comparing colony images of five microorganisms from different genera (Escherichia coli, Salmonella enterica, Pseudomonas aeruginosa, Staphylococcus aureus, and Candida albicans), the images showed obvious different features. Being closely related species, St. aureus and St. epidermidis resembled each other, but the imaging analysis could extract substantial information (colony fingerprints) including the morphological and physiological features, and linear discriminant analysis of the colony fingerprints distinguished these two species with 100% of accuracy. Because this system may offer many advantages such as high-throughput testing, lower costs, more compact equipment, and ease of automation, it holds promise for microbial detection and identification in various academic and industrial areas.

/pdf/colony-fingerprint-for-discrimination-of-microbial-species-2j1zj5vkep.pdf

Colony fingerprint for discrimination of microbial species based on lensless imaging of microcolonies.

The goal of this paper is to conduct the study on the text classification for financial news based on partial information. By a fact that an indispensable step for the efficient use of topic information embedded in financial news is the text classification, a new neural network called “All Dataset based on CharCNN (Character Convolutional Neural Networks) and GRU (Gated Recurrent Unit)” (in short, AD-CharCGNN) which extracts a part of the financial article and incorporates both time domain and spatial domain to classify financial texts is proposed. In the study of this paper, we first build a character level vocabulary by reading all characters of the financial dataset, part of each financial text which will be classified is mapped to a high-dimensional spatial vector based on the vocabulary. Then, the vectors are convoluted in the spatial domain to get the text local features, and next, the features are processed by the gated recurrent units to get the features contained time information. Finally, the features which contain spatial and time information will be classified through softmax function to get the text classification results. Our results on the experiments confirm that the network proposed in this paper works effectively with the accuracy of 96.45%, and it seems that the text classification algorithm with the feature by taking only partial text part is more suitable for the application of the practice. Meanwhile, for the input with character level vector, the network is not only suitable for Chinese but also for other languages.

/pdf/the-study-on-the-text-classification-for-financial-news-5bzlh6hboz.pdf

The Study on the Text Classification for Financial News Based on Partial Information

Neutrophils have a critical role in regulating the immune system. The immune system is compromised during chemotherapy, increasing infection risks and imposing a need for regular monitoring of neutrophil counts. Although commercial hematology analyzers are currently used in clinical practice for neutrophil counts, they are only available in clinics and hospitals, use large blood volumes, and are not available at the point of care (POC). Additionally, phlebotomy and blood processing require trained personnel, where patients are often admitted to hospitals when the infections are at late stage due to lack of frequent monitoring. Here, a reliable method is presented that selectively captures and quantifies white blood cells (WBCs) and neutrophils from a finger prick volume of whole blood by integrating microfluidics with high-resolution imaging algorithms. The platform is compact, portable, and easy to use. It captures and quantifies WBCs and neutrophils with high efficiency (>95%) and specificity (>95%) with an overall 4.2% bias compared to standard testing. The results from a small cohort of patients (N = 11 healthy, N = 5 lung and kidney cancer) present a unique disposable cell counter, demonstrating the ability of this tool to monitor neutrophil and WBC counts within clinical or in resource-constrained environments.

Monitoring Neutropenia for Cancer Patients at the Point of Care

/pdf/machine-learning-based-lens-free-shadow-imaging-technique-6c02j77g.pdf

Machine Learning Based Lens-Free Shadow Imaging Technique for Field-Portable Cytometry

Lens-free imaging technology has been extensively used recently for microparticle and biological cell analysis because of its high throughput, low cost, and simple and compact arrangement. However, this technology still lacks a dedicated and automated detection system. In this paper, we describe a custom-developed automated micro-object detection method for a lens-free imaging system. In our previous work (Roy et al.), we developed a lens-free imaging system using low-cost components. This system was used to generate and capture the diffraction patterns of micro-objects and a global threshold was used to locate the diffraction patterns. In this work we used the same setup to develop an improved automated detection and analysis algorithm based on adaptive threshold and clustering of signals. For this purpose images from the lens-free system were then used to understand the features and characteristics of the diffraction patterns of several types of samples. On the basis of this information, we custom-developed an automated algorithm for the lens-free imaging system. Next, all the lens-free images were processed using this custom-developed automated algorithm. The performance of this approach was evaluated by comparing the counting results with standard optical microscope results. We evaluated the counting results for polystyrene microbeads, red blood cells, and HepG2, HeLa, and MCF7 cells. The comparison shows good agreement between the systems, with a correlation coefficient of 0.91 and linearity slope of 0.877. We also evaluated the automated size profiles of the microparticle samples. This Wi-Fi-enabled lens-free imaging system, along with the dedicated software, possesses great potential for telemedicine applications in resource-limited settings.

https://www.mdpi.com/2075-4418/6/2/17/pdf

Automated Micro-Object Detection for Mobile Diagnostics Using Lens-Free Imaging Technology.

Conventional automatic document classification methods are currently faced with challenges in terms of learning time and computing power, owing to the ever-increasing amount of data on the web. In this paper, we propose an efficient classification method that uses time series-based dataset selection. In the proposed method, the dataset is split based on time series data and the best set of testing documents selected. The results of classification performance tests conducted using a Naive Bayes classifier indicate that using a small amount of data divided in terms of time series is more efficient than using the entire dataset for learning.

Experimental study of time series-based dataset selection for effective text classification

We present a method for detecting CD34, a blast marker required for leukemia diagnosis, using a lens-free shadow imaging system based on machine learning with 96% accuracy.

Machine Learning based CD34+ Cell Detection using Lens-free Shadow Imaging Technology

Federated authentication is a user authentication and authorization infrastructure that spans multiple security domains. Many overseas Web applications have been adopting SAML-based federated authentication. However, in Korea, it is difficult to apply the authentication because of the high market share of a specific Web (application) server, which is hard to use open-source SAML software and the high adoption of Java-based standard framework which is not easy to integrate with SAML library. This paper proposes the SAML4J, which is developed in order to have Web applications easily and safely integrated with the Java-based framework. SAML4J has a developer-friendly advantage of using a session storage independent of the framework and processing Web SSO flows through simple API. We evaluate the functionality, performance, and security of the SAML4J to demonstrate the high feasibility of it. 키워드 : SAML, SOAP, 자바, 스프링 프레임워크, 계정연합

http://koreascience.or.kr:80/article/JAKO201928462664796.pdf

Development of SAML Software for JAVA Web Applications in Korea

Videoconferencing has become a crucial enabler for sustainable collaboration and learning during the COVID-19 pandemic. However, national regulations often restrict public institutions from introducing commercial videoconferencing services. Open-source software is an attractive option for institutions if it can be protected from potential security threats while ensuring high usability. Unfortunately, to the best of our knowledge, we hardly find available open-source videoconferencing applications in the literature that stress their usability and adopt security-related frameworks. This study presents a federated-access management system called trustHub, which was developed to enable flexible and elaborate access control and protocol-agnostic user authentication. In addition, we introduce two videoconferencing applications that aim to improve the usability of leveraged open-source software. They are prototyped to operate in concert with trustHub to take firm access control and accept various types of identity providers. Consequently, using data collected from trustHub and a prototyped videoconferencing application over a 10-month period, we conduct a comprehensive analysis to understand the usage patterns of federated access and videoconferencing during the pandemic and, thus, verify their feasibility indirectly.

Yeonghun Chae

Papers

Automated Micro-Object Detection for Mobile Diagnostics Using Lens-Free Imaging Technology.

Experimental study of time series-based dataset selection for effective text classification

Machine Learning based CD34+ Cell Detection using Lens-free Shadow Imaging Technology

Development of SAML Software for JAVA Web Applications in Korea

Federated-Access Management System and Videoconferencing Applications: Results from a Pilot Service during COVID-19 Pandemic