1,720,997 research outputs found
IoT Recommender System: A Recommender System based on Sensors from the Internet of Things for points of interest
Get PDFThe Internet of things is a great source of knowledge. Its strength is based on the use of sensors that can provide information in real time on multitude of variables in any part of the world connected to an internet connection. At the same time, social networks increasingly encourage users to create comments on points of interest they visit or could visit, giving valuable feedback for other users or even for recommender systems that could use that information to anticipate user tastes in the future. In this work, we present an overview of a novel recommendation system to generate recommendations for users based on both the information gathered from sensors and the opinions explicitly indicated by users about places or recommendations previously made by the system
Automated human cortical bone Haversian canal histomorphometric comparison system
No full textThis research presents comparison of histological parameters of cortical bones for Malaysian population. For this research 44 samples were collected from Department of Pathology, Faculty of Medicine Universiti Kebangsaan Malaysia. Sample preparation being one of the important aspects of this research, is described in detail. Six Haversian canal-based parameters were selected for the comparison. Sex and race comparisons were performed on the collected Malaysian samples. Race comparison was performed between the two Malaysian races (Chinese-Indian). The results obtained from sex comparison demonstrates difference in mean Haversian canal area (hcm). The hcm parameter in female samples were found to be significantly larger than male samples (P < .05). This research also presents an automated system which can be used as a platform to perform Haversian canal parameter comparison. The system is designed to automatically check normality of the parameters measurement data and select the relevant comparison test. It is divided into two main section. The first section provides race comparison platform while the second section is dedicated for sex comparison
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
IJMLNCE Editorial Note Volume No 04, Issue No 02
No full textIt gives me pleasure to present the editorial preface of the International Journal of Machine Learning and Networked Collaborative Engineering (IJMLNCE) Volume 04 No 02 (2020). This issue comprises five manuscripts contributed by authors.
The first paper of this issue titles, "Edge detection of Friction Stir Welded Joints by using Fourier Transformation," is contributed by Akshansh Mishra et. Al. In the manuscript, the authors have implemented two machine learning-based image processing techniques. They stated that visual inspection had played a vital role in the beginning era of science. Nowadays, image processing is finding application for defects analysis of the manufactured parts in many industrial processes. We have implemented two machine learning-based image processing techniques in recent work, i.e., Fourier Transformation operator and Laplacian operator for the surface defects detection in Friction Stir Welded joints. In conclusion, The quality of the weld surface in the Friction Stir Welding process depends on the input parameters such as Tool Rotational Speed (rpm), Tool Traverse Speed (mm/min), and an Axial Force (kN).
The second paper, titled "Utilize Machine Learning Methods to Detect Plaintext Passwords," is authored by Nada Alnoaimi et. Al. In this article, information security explores, where the author states that every company is a target today, no matter its type. Hackers and cybercriminals are after data which they can monetize in many ways. Being proactive and have a defensive and protective plan in place, such as evaluating and assessing IT security, is an excellent recipe for avoiding data breaches and, consequently, business disasters. Why not utilizing a machine learning platform could be trained to search text in a computer resource, detect a string of plaintext characters, and analyze the string of characters to predict or detect a plaintext password on a computer resource asset. The machine will be able to catch a plaintext password in a character string by analyzing plaintext character strings for typical password complexity, such as, for example, including at least one uppercase letter, lowercase letter, number, unique character, and text length (for example, minimum of eight characters). It will also predict a level of certainty that a character string includes a password and output a confidence score based on the expected level of certainty. Finally, it will categorize the confidence score in any number of prediction certainty levels, including, for example, three groups – high, medium, or low.
S Nagaprasad et al. contributed the third article for this issue titled "Heart Disease Prediction Propagation approach." Data mining methods are used to test complex data, and regression processing based on input data sets is used to estimate results. A variety of prediction analysis methods have been implemented in recent years. The clustering method k-means and SVM ( support vector machine) are a statistical, computational technique for clustering and defining primary data to detect cardiac disorders. In this study, the Back Propagation Method is used in tandem with the k-means clustering algorithm to cluster knowledge for improved prediction research performance. The implemented algorithm\u27s output is found in the cardiac disorder data sample collected from the UCI depositor. Within this sample, there are 66 attributes.Nonetheless, a subgroup of 14 qualities is needed for every study. The Cleveland platform is utilized in particular for machine-learning investigators. The research designed correlates with the current techniques, precision, error identification, and deployment time (using the numerical mean).
The Fourth article, title "Discovering Trending Topics from the Tweets By Odia News Media During Covid-19," was contributed by Swarupananda Bissoyi et. Al. This paper explores the Covid-19 pandemic\u27s onset, and the lockdown imposed because it has significantly fueled news consumption. News portals, including the ones in Odia language, are actively feeding news related to Covid-19 to their consumers via their websites and Twitter handles. The news items didn\u27t restrict to Covid-19 alone; they also touched various domains of life like education, healthcare, administration, politics, movies, etc. Discovery of the news trends provides a bird\u27s eye view of the issues and topics popular in the online community. This could be of interest to advertisers, marketers, researchers, sociologists, and policymakers. This paper applies Topic Modeling to discover the trends from the tweets made by the Odia news media from 20th March 2020 to 31st August 2020, the period which saw the emergence of both lockdowns and unlocks in India. We found that during this period, the Odia news media didn\u27t restrict themselves to report news surrounding Covid-19; rather they reported other happenings as well.
In the fifth and last article of this issue, titled "Designing Hand-Held Vibration Measuring Device for Industrial Machines," is contributed by Thi Dieu Linh Nguyen et al., In this manuscript, the authors discuss that Evaluating the quality of industrial machines, the vibration meter is used to measure the actual vibration of the machine. The two most important parameters describing machine vibration, amplitude, and frequency, are the basis for determining the cause of vibrations. Spectral analysis of the vibration signal will give information about the vibration level and choose which part of the machine the vibration signal is caused. This paper presents the manufacturing of vibration measuring devices with simple structure, compact size, and high accuracy at a reasonable price. The spectral analysis method of vibrating signals and real-time spectrum display of the measured vibration signals
I am sure that these five papers included in the International Journal of Machine Learning and Networked Collaborative Engineering (IJMLNCE) Volume 04 No 02 (2020) will be useful to the research community. At this end, I am thankful to the Editorial board member for their timely support in the review. I am looking forward to receiving your unpublished research work for Volume 05 No 01 (2021)
IJMLNCE Editorial Note Volume No 04, Issue No 01
No full textIt gives me pleasure to present the editorial preface of the International Journal of Machine Learning and Networked Collaborative Engineering (IJMLNCE) Volume 04 No 01 (2020). This issue comprises five manuscripts contributed by authors.
The first paper contributed by Ramachandran A et al. titles, "Efficient Distributed Web Crawler Using Hefty and Enhanced Bandwidth Algorithms for Drug Website Search," In this paper, a novel Hefty Algorithm and enhanced bandwidth algorithm are combined for a better-distributed crawling system. The hefty algorithm was implemented to provide efficient and robust surfing results while applying on the drug web search. Refabricate a proficient search structure is very important due to the current scale of the web. Search engines mine information from the web, and a web crawler program that surfs the web in an efficient manner. A distributed crawler belongs to a variant of a web crawler, uses a dispersed computation method. In this paper, we design and implement the concept of Efficient Distributed Web Crawler using enhanced bandwidth and hefty algorithms. Mostly Web Crawler does not have any distributed cluster performance system and any implemented algorithm. We also implemented an Enhanced Bandwidth algorithm to improve the efficiency of the proposed crawler.
The second paper, titled "Revealing Brain Tumor Using Cross-Validated NGBoost Classifier," is authored by Shawni Dutta and Samir Bandyopadhyay. In this article, the author writes that the brain is the most complicated and delicate anatomical human body structure. Statistics prove that, among various brain ailments, brain tumors are most fatal, and in many cases, they become carcinogenic. Brain tumors are characterized by abnormal and uncontrolled growth of brain cells. It takes up space within the cranial cavity and varies in shape, size, position, and characteristics viz. It can be benign or malignant, which makes the detection of brain tumors very critical and challenging. A neurologist or neurosurgeons vital information needs to have is the precise size and location of cancer in the brain and whether it is causing any swelling or compression of the brain that may need urgent attention. This paper exploits ensemble strategy-based Machine Learning (ML) algorithms for revealing brain tumors. NGBoost algorithm and 5-fold stratified cross-validation scheme are proposed as classifier models that automatically detect patients with brain tumors. The proposed method is implemented with necessary fine-tuning parameters compared against ensemble-based baseline classifiers such as AdaBoost, Gradient Boost, Random Forest, and Extra Trees Classifier. An experimental study implies that the proposed method outperforms baseline models with significantly improved efficiency. The interfering features that impact brain tumor classification are ranked, and this ranking is retrieved from the best classifier model.
Abdulmohsen Alotaibi contributed the third article for this issue titled "Transfer Learning for Detecting Covid-19 Cases Using Chest X-Ray Images". In this interesting work, he is discussing the COVID-19 pandemic, which is a global health crisis that has already infected more than 3.5 million people and caused more than 250 thousand deaths around the globe. He focused on developing a more efficient way to detect and treat this illness. This paper utilizes transfer learning techniques to detect normal, COVID-19, and viral pneumonia cases from Chest X-Ray images. Four pre-trained models on ImageNet were chosen as the base model, which are ResNet50, VGG19, DenseNet121, and InceptionV3. The performance metrics of each fine-tuned model are overall similar. With an average recall, precision, f1-score, and accuracy of 97.42%, 97.42%, 97.23%, 98.3% respectively.
The Fourth article, title " The Economic Impact of Social Media Fraud and it\u27s Remedies," was contributed by shakik mahmud et al. This paper presents the economic impact of social media fraud in Bangladesh and its IT-based prevention model. Online privacy and security problems become a big concern online day by day—many types of issues growing up here, for example, phishing, hacking, sabotage, etc. Social media is a popular and powerful tool to express personal life and also business purposes in Bangladesh. Social communicating websites such as Facebook, Twitter, WhatsApp, and LinkedIn are popular social sites. Facebook is the most popular one. Through these media, people communicate with their friends and family and share thoughts, photos, videos, and lots of data. Many types of business and commerce have developed on social media. Presently, people depend on it, so it\u27s marketing value increases day by day well. Some Tech fraud groups have been formed and wake up to hack money in some tricky way in this big virtual society. At present, social media is one of the critical areas for fraudsters. This paper will show based on our study how much money is being spent through it and others by the IT-based prevention model of this problem.
The fifth article of this issue, titled " The Performance Enhancement Systems of Human Iris Pattern and Recognition Method through Digital Authentication Application," is contributed by Krishnaveni N et al. Authors discusses that human iris and recognition patterns have been recognized as the best biometric marking ever found. The iris and the textured iris patterns\u27 uniqueness tend to remain natural, unchangeable, and recognizable through existence. Mathematical analyses of the unique stable patterns formed within the iris include Iris detection methods, and a comparative analysis is carried out utilizing an established database. In this document, a clean electoral system is created to build a fraud-free ID list of electors. To find the Iris and Eyes, the algorithm of canny edge detection is used, Dougman\u27s normalization procedure is used, object filters are added, and the corresponding process is finally conducted for the Euclidian set. Biometric authentication confirms our identification by being a simple and increasingly secure method. They implemented a weighted, majority voting process for all biometric authentication systems utilizing a bitwise contrast between inscription and biometric models to resolve this problem and enable Iris identification in less than ideal images. We also observed that the approach outdoes the current majority and efficient bit sorting strategies through a set of tests with the database CASIA iris. Our process is an easy and efficient way to boost the accuracy of established iris detection systems.
I am sure that these five papers included in the International Journal of Machine Learning and Networked Collaborative Engineering (IJMLNCE) Volume 04 No 01 (2020) will be useful to the readers and researchers. At this end, I am thankful to the Editorial board member for their timely support in the review. I am looking forward to receiving your unpublished research work for Volume 04 No 04 (2020)
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Get PDFWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
IJMLNCE Editorial Note Volume No 03, Issue No 02
No full textThe International Journal of Machine Learning and Networked Collaborative Engineering (IJMLNCE) with ISSN: 2581-3242is now indexed in popular databasessuch asBASE (Bielefeld Academic Search Engine), CNKI Scholar, CrossRef, CiteFactor, Dimensions, DRJI, Google Scholar, Index Copernicus, JournalTOCs, J-Gate, Microsoft Academic, PKP-Index, Portico, ROAD, Scilit, Semantic Scholar, Socolar or WorldCat-OCLC.We are now proud to present the eighth volume of the journal, Volume No-03 Issue No-02, with some high-qualitypapers written by international authors and covering different aspects related to machine learning and collaborative engineering.
Puri et al. published a work entitled “Cloudbin: Internet of Things based waste monitoring system”.In this paper, authors present an IoT-based waste management system called Cloudbin to monitor and control waste garbage in urban areas. To that end, authors use different elements like an ultrasonic sensor, a GPS module or a methane detection mechanism. The problem of waste management is one of the key elements in which governments must take an active part.
Rimal published a work entitled “Machine Learning Prediction of Wikipedia Time Series Data using: R Programming”. In this work, author explains how prediction of automatic learning of Wikipedia time series work using the R environment. To that end, author focused on real data from Cristina Ronaldo, a famous football player, presenting, according to the author, the simplest way to predict times series data and its strengths for data analysis.
Sen et al. published a work entitled “Study of Energy Efficient Algorithms for Cloud Computing based on Virtual Machine Migration Techniques”. This study describes how energy efficiency in cloud computing is one of the most important features to be considered to measure the efficiency of such services, balancing power and quality of the service. Thus, authors discuss how virtual machine migration techniques can help to achieve energy efficiency.
Choudhary published a work entitled “Information Processing in GLIF Neuron Model with Noisy Conductance”. Authors investigate the generalized leaky integrate-and-fire neuron model with stochastic synaptic conductance and investigate the effect of varying concentration of electro-chemicals at the synapse in a single neuron model. To that, they developed a simulation-based study with the temporal encoding technique to analyze the encoding mechanism.
Finally, Kothandan and Sujatha published a work entitled “Deep Neural Network with Stacked Denoise Auto Encoder for Phishing Detection”. In this paper, authors present and discuss a deep neural network to detect phishing uniform resource locators. They use a feature vector with a stacked denoise auto encoder. In addition, the noisy data is trained to reconstruct a clean input feature vector. Experiments are based on the Ham, Phishing Corpus and Phishload datasets to prove its effectiveness
Volume No 04 Issue No 03 (2020)
No full textFace Recognition Approach using Stereo Matching Algorithm
Enhancing the Accuracy of Indoor Positioning Using System Delay Time Compensation
E-Recruitment In HR Consultants via E-Technology
A Model on Fuzzy Logic Implementation in the Development of Traffic Management in Smart Cities: Artificial Intelligence Approach
IoT and AI-based plant monitoring syste
- …
