International Journal of Intelligence and Sustainable Computing (9 papers in press)

Regular Issues

• Latent Semantic Analysis in Automatic Text Summarization: A state of the art analysis  
  by Tapas Guha, Mehala N 
  Abstract: Increasing availability of information in the web and its ease of access necessitates the need for efficient and effective automatic text summarization. Automatic text summarization condenses the source text (a single document or multiple documents) into a compact version preserving its overall meaning and information content. Till now, researchers have employed different approaches for creating well-formed summaries. One of the most popular methods is the Latent Semantic Analysis (LSA). In this paper, various prominent works to produce extractive and abstractive text summaries based on different variants of LSA algorithm are reviewed, analysed and compared.
  Keywords: Information retrieval; Automatic Text Summarization; Latent Semantic Analysis; Singular Value Decomposition.
  DOI: 10.1504/IJISC.2020.10029282
   
  
• Ensemble of Artificial Neural Networks and K-Nearest Neighbor for classification of granite images  
  by Fisha Haileslassie 
  Abstract: This study attempted to develop a granite quality classification model by comparing colour, texture and ensemble of colour and texture Image processing technique based on combined texture and colour features was examined to achieve good results An average of 120 pictures taken for each granite (grade A, grade B, grade C) A grayscale coexistence matrix used for texture and colour histogram for colour extraction Five textures and six colour features were extracted from each granite To build the models for the prediction K-NN, ANN and Ensemble of KNN and ANN are examined Experimental results, the ANN and KNN ensemble model shows good outcomes with combined texture and colour features using SFFS methods An average accuracy of 85 3%, 93 6% and 95 8% is achieved for KNN, ANN and ensemble of KNN and ANN Granite fractures and vines of the images have a strong impact on the performance of the classifier.
  Keywords: Classification of granite; feature extraction; ANN; K-NN; an ensemble of ANN and K-NN.
  DOI: 10.1504/IJISC.2020.10030717
   
  
• Emotional intelligence and Performance of Faculty in Knowledge Engineering Education  
  by Tenreiro Machado 
  Abstract: Faculty are instrumental in moulding and nurturing global citizens of future. Knowledge engineering looks very fascinating in terms of a rewarding career ahead and management education institutions are striving to not only to compete with each other but also to survive in current times of a dull employment scenario. There has been a drastic fall at the input level in most institutions that are affiliated to Universities and consequently the level of placements has suffered drastically with only a bright few getting decent salaries. The current scenario in education system requires herculean efforts on the part of the Management and especially the main facilitators, namely the faculty, to provide all resources and training to mould the students and transform them to be accepted by Corporate. Research was undertaken to analyse the levels of intelligence for the faculty alongside its effects on the performance given the fact that faculty are the core resource of such higher education institutions.
  Keywords: Knowledge engineering; post-graduation; institutions; emotional intelligence; faculty performance.
  DOI: 10.1504/IJISC.2020.10030890
   
  
• SMI Attributes: Key role in business as a service in Cloud Computing  
  by Thasni Thaha, Kalaiarasan C 
  Abstract: The development of cloud computing leads to a situation that, unique services are being offered by an ever increasing number of organisations. It is very difficult for the customer to choose the right cloud provider based on his service requirements. There is no framework to index cloud provider that depends on the user requirement. Undertakings from new businesses to huge organisations see cloud as a decent choice, in light of asset pooling, on interest provisioning and pay as you go model. The most essential objective for a cloud specialist organisation is making a feeling of certainty and trust about that cloud provider among service requesters. To accomplish this, associations ought to build up a technique that is complete, efficient and it ought to be finished with legitimate checks understanding the future necessities, and by conveying a cloud framework. This review paper centres around different characteristics of cloud benefit estimation record created by Cloud Services Measurement Initiative Consortium (CSMIC) that gives a standard technique to estimating and looking at a business benefit.
  Keywords: cloud; service provider; SMICloud; trust; Service Level Agreement; SLA; quality of services; QoS.
  DOI: 10.1504/IJISC.2020.10031250
   
  
• REAL TIME PEOPLE FLOW COUNTER AND ESTIMATION USING DEEP LEARNING  
  by Anand R 
  Abstract: A real-time people flow counter and estimation using deep learning is the significant device in computer vision and a predictable knowledge discovery application in security, entertainment, tourism and corporate business. However, the state-of-the-art machine learning, deep learning and computer vision methods have complete this technology as a game-altering and even better human identification and counting part in terms of accurateness. This paper focuses on one of the progressive deep learning tools in people counting to achieve higher efficiency. Also, focus on Image processing approach to count the number of people entering and exiting a defined place using neural networks. Here, the model is trained by artificial neural networks and computer vision. Real-time people flow estimation is very important for several applications like security, business, tourism, and other fields where people flow surveillance is required.
  Keywords: centroid tracker; human detection; artificial intelligence; deep learning; OpenCV; neural network.
  DOI: 10.1504/IJISC.2020.10031301
   
  
• Aggressive Forward Infinite Dimensional Entangled State Pair With Reputation Based Weighted Hub Selection To Packet Loss Avoidance  
  by R. Regin, J. Murali, S.Pradeep Kumar, C. Karthik, C. Kasilingam 
  Abstract: Mobile Ad Hoc Network (MANET) is indeed a series of nodes which have been put with others to collaborate without infrastructure MANET needs to constantly relay on wireless communication because of its limited characteristics It contains a mobile wireless transmission node and MANETs which can be set anywhere Every component of the ad-hoc network therefore is effective because it can reduce the burden of central admission and infrastructure Across many ways, the MANET is widely used, including military applications, mobile phones, and crises In this Paper first proposes an Aggressive Forward Infinite-dimensional Entangled Pair state (AFIDEPS) that initiates Pair stateby allocating the key to the server and routing the key to that node afterwards When mapping keys into the hubs, area administration and traditional name provides direct mapping of values from keys The Second is the Reputation based weighted Hub Selection (RWHS) Route selection seems to be the process of
  Keywords: Pair state; Hub; Reputation; Infinite-Dimensional.
  DOI: 10.1504/IJISC.2020.10031372
   
  
• Smart Personalized Recommendation System for Wanderer using Prediction Analysis  
  by L. Maria Michael Visuwasam, M. Geetha, G. Gayathri, K. Divya, D. Elakkiya 
  Abstract: Indian Tourism is one of the Rapid Growing Industry, plays one part in country economies. As it is not easy to get smart suggestions for place and accommodation to personalized travel itineraries or both individual and group of tourists based on the interest preference. The work aims to propose a platform for extract and give output based on user point of interest in tourism. Extraction of opinions from user reviews, specific to accommodation services, are useful to the clients looking for accommodation. The proposed system extracts the famous place reviews using the tags and comments from Flicker (public website). Dataset then we classify them, using the POI Algorithm technique. The recommendation system also helps to solve problems by providing the Budget estimation using the cost estimation Algorithm in Machine learning.
  Keywords: POI Algorithm; Machine Learning; Cost Estimation; Map API; weather forecasting; Seasonal extraction; Recommendation system.
  DOI: 10.1504/IJISC.2020.10031883
   
  
• 1D CNN Architectures for EEG Classification with Motor Imagery input of eyes open and eyes closed conditions  
  by RADHA SUBRAMANYAM, NAGABUSHANAM PERATTUR, S. THOMAS GEORGE, SUBATHRA M. S. P 
  Abstract: Neuro cognitive performance is an interesting area which deals with normal and abnormal conditions of a person in his thinking capabilities like driving, reaction to the events happening around him. The study of such information is called sleep scoring or sleep stages classification. If it is abnormal for a person, it further leads to insomnia, hypersomnia, epilepsy disorders. Hence, early stage detection of such abnormalities may help in treating the person intime and avoiding major problems ahead. Sleep stage classification may be 2-way, 3-way, 4-way or 5-way based on type of EEG captured from the person. In this paper, we have proposed various 1D-CNN architectures for 2-way sleep stage classification for motor imagery EEG captured during eyes open and eyes closed conditions. Implementation is done in python keras in Jupiter notebook, anaconda. Among the proposed models, 1D-CNN with 4 layers give better accuracy of 75.753%. In general, sleep stages classification give accuracy less than 80% and also it may be based on quality of EEG input considered.
  Keywords: 1D-CNN; motor imagery EEG; classification; trainable parameters; convolutional layers.
  DOI: 10.1504/IJISC.2020.10032562
   
  
• Design and Analysis of dual feed patch antenna at 3.5 GHz Mid-band 5G Technology  
  by Prabhu T 
  Abstract: A circularly polarized (CP) patch antenna with dual feed is presented. The antenna consists of a radiating patch, substrate and ground plane. A pair of feed is placed at TM10 and TM01 mode with same amplitude and 900 phase differences. The two orthogonal modes are excited by using coaxial probe feed. By changing the phase difference of two modes, both the left and right hand circularly polarized waves are obtained. The antenna is designed at mid-band 5G technology (i.e 3.5 GHz) and simulated results are obtained with return loss of -25.03 dB, impedance bandwidth of 74.6 MHz and gain of 7.87 dBic. The proposed antenna had been considered as an excellent option for mid band 5G Technologies to improve the coverage and capacity benefits.
  Keywords: patch antenna; 3.5 GHz; 5G technology; dual-feed; bandwidth.
  DOI: 10.1504/IJISC.2020.10033067