International Journal of Machine Learning and Networked Collaborative Engineering (IJMLNCE)
Not a member yet
85 research outputs found
Sort by
IJMLNCE Editorial Note Volume No 03, Issue No 01
No full text
The International Journal of Machine Learning and Networked Collaborative Engineering (IJMLNCE) with ISSN: 2581-3242 continues its growth. The journal is becoming more and better indexed in platforms such as BASE (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, WorldCat-OCLC. After more than two years of intense work, we are proud to present the seventh volume of the journal, Volume No-03 Issue No-01, which introduces five high quality works written by recognized authors that deal with different aspects within the scope of the journal.
Nigar published a work entitled “A Study on Internet of Things in Women and Children Healthcare”. Author focuses on the importance of the new internet of things related- technologies, which can be applied to healthcare. In fact, its integration with electronic health and telemedicine is gaining attention. The paper describes some methods, practices and prototypes based on the internet of things in the field of healthcare focusing on women and children.
Gunagweare and Kiani published a work entitled “Ultimate Indoor Navigation: A Low Cost Indoor Positioning and Intelligent Path Finding”. Authors deal with the drawbacks of the global positioning system (GPS), which is not useful in indoor environments or places where some buildings can interfere with the satellite signal. In this paper, authors present a simple, low-cost, context-aware and user-friendly indoor navigation system based on a common smart phone.
Jaidev et al. published a work entitled “Artificial Intelligence to Prevent Road Accidents”. Authors focus on the traffic congestion that in turn can lead to more car accidents. The idea of the authors is to study and review the literature related to approaches for detecting unsafe driving patterns to predict accidents with the help of artificial intelligence. Two apparently similar but different examples could be drivers under the influence of drugs or drivers under the influence of alcohol.
Rimal published a work entitled “Deterministic Machine Learning Cluster Analysis of Research Data: using R Programming”. The paper discusses various types of cluster analysis of different data sets with large number of dimensions (iris, utilities, mclust and dbscan). The main goal is to explain the simplest way for clustering analysis whose data structure is wide scattered. The work of the author is based on the R programming language and several specific packages.
Gunjal and Shaik published a work entitled “A Robust Decomposition Based Algorithm for Removal of Pattern Noise from Images”. Authors work on a melting pool of complex vectors to present a technique that requires less computer resources and less time for any image removal of pattern noise compared to other previously stated strategies. The work is based on the idea a picture includes components that can be described separately.
 
Cloudbin: Internet of Things Based Waste Monitoring System
No full textNowadays, waste management has become a critical issue for the environment. Government and private agencies need to take certain action for proper management and cleanliness. The absence of systematic waste management system creates many issues for the environment and living creatures. Research on the Internet of Things (IoT) applications widely increased in many sectors. The waste management system is also one of the sectors. Therefore, in this study, IoT based waste monitoring system called Cloudbin is proposed to reduce the waste garbage from urban areas. In this system, Ultrasonic sensor is fixed on the top of the waste bin to monitor the level of garbage inside the bin and connected to the Blynk server. In addition, a GPS module is also employed to check the location of Waste Bin. Methane detection from garbage is an important feature in the system. Results show that the proposed system is suitable to monitor and control waste in cities. 
Artificial Intelligence to Prevent Road Accidents
No full textDue to increasing demand in urban mobility and modern logistics sector, the vehicle population has been growing progressively over the past several decades. A natural consequence of the vehicle population growth is the increase in traffic congestion which in turn will lead to more accidents. Accident prediction is one of the most vital aspects of road safety. An accident can be predicted before it occurs, and precautionary measures can be taken to avoid it. Artificial Intelligence (AI) can help in improved awareness of road conditions, driving behaviour of the people and can avoid accidents with the help of improved active safety and improved traffic condition. Drug impaired driving is becoming a serious cause of accidents as the days go by. Moreover, it is more difficult to detect drivers who are under the influence of drugs than drivers who are the influence of alcohol. So the purpose of this research is to study and review the literature & industry reports and put in the approaches for detecting the unsafe driving pattern and also maintaining the health of the car to avoid accidents
Study of Energy Efficient Algorithms for Cloud Computing based on Virtual Machine Migration Techniques
No full textGreen cloud is a catchphrase in today’s IT industry and hence energy efficiency in cloud computing is one of the most significant parameters to follow nowadays to evaluate the efficiency of the cloud service. It is a driving force for adaptability of a cloud computing service in recent era. For a highly commercial service like cloud, maintaining the QoS parameters and keeping the service availability and service quality highly optimized to get the competitive advantage, cloud data centers are almost available on a 24x7 basis ; which in turn is a reson for high power consumption. So it is very much necessary to maintain a balance between power and quality of the service. One feasible solution for achieving energy efficiency is Virtual Machine migration technique in real time or when they are in turned off condition. This paper discusses about several VM Migration techniques and analyses their perspectives
Joint Spatial Geometric and Max-margin Classifier Constraints for Facial Expression Recognition Using Nonnegative Matrix Factorization
No full textBased on the constrained non-negative matrix factor algorithm, the article presents a new approach to facial recognition recognition. Our proposed method incorporated two tasks in an automatic expression analysis system: facial feature extraction and classification into expressions. To obtain local and geometric structure information in the data as much as possible, we amalgamate max-margin relegation into the constrained NMF optimization, resulting in a multiplicative updating algorithm is additionally proposed for solving optimization quandary. Experimental results on JAFFE dataset demonstrate that the effectiveness of the proposed method with improved performances over the conventional dimension reduction methods
Knowledge Graph-based Recommendation Systems: The State-of-the-art and Some Future Directions
No full textThe unprecedented growth of unstructured data poses many challenges in semantic computing, which is an active research area for many years. While unearthing interesting patterns such as entities, relationships, and other metadata are important, it is equally important to represent them in an efficient, easy to access manner. Knowledge Graphs (KGs) are one such mechanism to represent facts extracted from unstructured text. KGs represent entities as nodes and relationships as edges. Such a representation may find applications in many meaning-aware computing applications such as question answering, summarization, etc., to name a few. Very recently, knowledge graph-based recommendation systems have become popular which has many advantages over traditional recommendation engines. This survey is an attempt to summarize and critically evaluate some of the very recent approaches to knowledge graph-based recommendation approaches
Understanding Machine Learning for Friction Stir Welding Technology
No full textMachine learning process drastically decreases the time it takes to develop stronger, lighter materials. This is important to the aerospace, automotive and manufacturing sectors. Machine Learning techniques like Artificial Neural Networks and Image processing are used in Friction Stir Welding process for the optimization of mechanical properties like Ultimate Tensile Strength, Fracture Strength and elongation % and microstructure properties like grain size and understanding defects formation. In the recent paper, application of Machine Learning technique in Friction Stir Welding technology will be discussed
A Robust Decomposition Based Algorithm For Removal Of Pattern Noise From Images
No full textThis article aims a melting pool of complex vectors, that is, the aggregation and the minimization problem of sufficiency spectra. A mixture of this blended standard and image decline issue works admirably to reduce and deteriorate the example of concussion which occurs when old pictures are filtered with granular surfaces. In most cases, the appealing appropriation of regular photos easily reduces from low repetition to the high repetition band, while the episode of concussion is scarcely circulating. We agree along these lines that a picture viewed includes an idle image and an example clamor, describing them separately by using the full range and capacity work. This enables the two parts to decompose sensibly. In contrast to the comparative strategies of deterioration, for instance, robust PCA, our technique is decent, less computer expenditure, and moreover less time suited for any image organizatio
Deterministic Machine Learning Cluster Analysis of Research Data: using R Programming
No full textThis review paper clearly discusses the compression between various types of cluster analysis of different data sets were explained sufficiently. Although there is large gap between the way of analysis of collected data and its cluster categorization research data using r programming. Its primary purpose is to explain the simplest way of clustering analysis whose data structure were wide scattered using R software whose outputs were sufficiently explain with various inter-mediate output and graphical interpretation to reach the conclusion of analysis. Therefore, this paper presents easiest way of clustering when data sets with large dimensions with multivariate analysis and its strengths for data analysis using R programming
Deep Neural Network with Stacked Denoise Auto Encoder for Phishing Detection
No full textSensitive information such as credit card information, username, password and social security number etc, can be stolen using a fake page that imitates trusted website is called phishing. The attacker designs a similar webpage either by copying or making small manipulation to the legitimate page so that the online user cannot distinguish the legitimate and fake websites. A Deep Neural Network (DNN) was introduced to detect the phishing Uniform Resource Locator (URL). Initially, a 30-dimension feature vector was constructed based on URL-based features, Hypertext Markup Language (HTML)-based features and domain-based features. These features were processed in DNN to detect the phishing URL. However, the irrelevant, redundant and noisy features in the dataset increase the complexity of DNN classifier. So the feature selection is required for efficient phishing attack detection. But feature selection is a time-consuming process since it is an independent process. So in this paper, a feature vector is generated by DNN itself using Stacked Denoise Auto Encoder (SDAE). Moreover, the noisy data such as missing features affect the efficiency of phishing detection so the SDAE is trained to reconstruct a clean input feature vector. The initial input feature vector is corrupted by setting some feature vectors as zero. Then the corrupted feature vector is then mapped with basic auto encoder, to a hidden representation from which the input feature vector is reconstructed. The reconstructed features are given as input to DNN which selects the most relevant features and predicts the phishing URL. Hence the sparse feature representation of SDAE increases the classification accuracy of DNN. The experiments are conducted in Ham, Phishing Corpus and Phishload datasets in terms of accuracy, precision, recall and F-measure to prove the effectiveness of DNN-SDAE