76 research outputs found
Malicious url classification using machine learning algorithms and comparative analysis
No full textExponential expansion in the application of the internet in each and every field has resulted in the escalation of data traffic over the internet. In the vastness of this data it has become important for engineers to classify the data as malicious and non-malicious so that different traffic can be treated differently. Rule-based and port-based classification exhibited a number of limitations which ultimately led to the steep decline in their usage to classify the internet traffic and gave rise to the machine learning techniques which are more promising and efficient. In this paper four popularly known machine learning classifiers: KNN, Naive Bayes, Decision Trees and Random forest have been implemented to classify the internet traffic based on whether the traffic is malicious or not and then compare their results on the basis of their accuracy score
Document Clustering with Map Reduce using Hadoop Framework
Get PDFBig data is a collection of data sets. It is so enormous and complex that it becomes difficult to processes and analyse using normal database management tools or traditional data processing applications. Big data is having many challenges. The main problem of the big data is store and retrieve of the data from the search engines. Document data is also growing rapidly in the eon of internet. Analysing document data is very important for many applications. Document clustering is the one of the important technique to analyse the document data. It has many applications like organizing large document collection, finding similar documents, recommendation system, duplicate content detection, search optimization. This work is motivated by the reorganization of the need for a well efficient retrieve of the data from massive resources of data repository through the search engines. In this work mainly focused on document clustering for collection of documents in efficient manner using with MapReduce.
DOI: 10.17762/ijritcc2321-8169.15018
Machine Learning Approach to Improve Data Connectivity in Text-based Personality Prediction using Multiple Data Sources Mapping
No full text109-119This paper considers the task of personality prediction using social media text data. Personality datasets with
conventional personality labels are few, and collecting them is challenging due to privacy concerns and the high expense of
hiring expert psychologists to label them. Pertaining to a smaller number of labelled samples available, existing studies
usually adds a sentiment, statistical NLP features to the text data to improve the accuracy of the personality detection model.
To overcome these concerns, this research proposes a new methodology to generate a large amount of labelled data that can
be used by deep learning algorithms. The model has three components: general data representation, data mapping and
classification. The model applies Personality correlation descriptors to incorporate correlation information and further use
this information in generating dataset mapping algorithm. Experimental results clearly demonstrate that the proposed
method beats strong baselines across a variety of evaluation metrics. The results had the highest accuracy of 86.24% and
0.915 F1 measure score on the combined MBTI and Essays dataset. Moreover, the new dataset constructed contains
3,84,089 labelled samples on the combined dataset and can be further considered for personality prediction using the famous
Five Factor Model thereby alleviating the problem of limited labelled samples for the purpose of personality detection
Fatigue behaviour of FDM-3D printed polymers, polymeric composites and architected cellular materials
Get PDFPolymer-based materials are increasingly produced through fused deposition modelling (FDM) – an additive manufacturing process, due to its intrinsic advantages in manufacturing complex shapes and structures at low overhead costs. The versatility of this technology has attracted several industries to print complex geometrical structures. This underlines the importance of studying the mechanical strength of FDM printed polymeric materials, especially their fatigue behaviour in cyclic loading conditions. Conventionally manufactured polymeric materials (e.g. injection moulding) have superior fatigue performance than FDM printed materials. Unlike conventionally manufactured polymers, FDM-made polymers have layer by layer adhesion and the influence of printing parameters make fatigue analysis complex and critical. The influences of printing parameters and printing material characteristics have a significant impact on the fatigue behaviour of these materials. The underlying mechanism behind the fatigue of FDM printed polymers is crucial for the assessment of these materials in structural applications. However, the fatigue behaviour of FDM printed polymeric materials has not been reviewed in detail. Therefore, this article aims to evaluate 3D printed polymeric materials’ fatigue properties. The importance of fatigue in the FDM printed biomedical materials is also reviewed, and more importantly, the novel FDM printed architected cellular material fatigue properties are also introduced. © 2020 The Author(s
Hydrogenated amorphous silicon solar cells fabricated at low substrate temperature 110°C on flexible PET substrate
No full textPartition Based Clustering Using Genetic Algorithm and Teaching Learning Based Optimization: Performance Analysis
No full text
Effect of dilute magnetic ions on the optical, dielectric and ferroelectric properties of PZT at morphotopic phase boundary
No full textA Comparative Study on Effective Approaches for Unsupervised Statistical Machine Translation
No full text
- …
