Hybrid spam detection using machine learning
Social networks are recognized as popular communication channel but in this, there is one of the problem is spam messages. Spam messages can contain malware in the form of the executable file and the link to the malicious websites or the links which do not exist. Most of the existing machine learning solutions are based on the either Support Vector Machine or Naive Bayes but the existing solutions either slow or inaccurate in solving spam filtering problem. Support Vector Machine based spam filter has great advantages on high precision and recall rate and Naive Bayes based spam filter give faster classification speed and require small training sets. By taking the advantages of both, we propose hybrid spam filtering algorithm which have more accuracy than separately implemented NB and SVM.
Published by: Diksha S. Jawale, Ashwini G. Mahajan, Kalyani R. Shinkar, Vaishnavi V. Katdare
Author: Diksha S. Jawale
Paper ID: V4I2-1925
Paper Status: published
Published: May 1, 2018
Applied to animal breeding for disease resistance
Animal diseases cause significant losses to livestock’s and breeders, resultant into direct damages to animals, reduced productivity and cost of treatment. One of the promising additional methods to control diseases is breeding animals for disease resistance. The resistance of animals to diseases is related to their ability to withstand pathogens and harmful environmental influences. For the planned control of hereditary diseases in domestic animals, it is required not only an accurate knowledge of their inheritance but also an accurate and monotonous description and designation of the disease. Therefore, in order to avoid misunderstandings in identifying hereditary anomalies in domestic animals, an international nomenclature has been created for their description and designation. According to this nomenclature, hereditary diseases of each species of animals are designated by a certain letter of the alphabet. Specific diseases peculiar to this species are characterized by numerical indexes. Developments in technologies, such as genomic selection, may help overcome several of the limitations of traditional breeding programmes and will be especially beneficial in breeding for lowly heritable disease traits that only manifest themselves following exposure to pathogens or environmental stressors in adulthood. The current paper provides a brief overview of the present-day application of microsatellites markers in animal breeding and makes a significant contribution to the overall farm animal health and resistance to disease.
Published by: Ashraf Ward, Ismail M. Hdud, Omry M. Abuargob, Sergey Y. Ruban
Author: Ashraf Ward
Paper ID: V4I3-1204
Paper Status: published
Published: May 1, 2018
Enhancing cloud security by analyzing venerability of cloud server in DDOS attack
Now a days more and more services and applications are emerging in the Internet, exposing sensitive electronic data in the internet has become easier. Web services cause’s personal data to be cached, copied, and archived by third parties, often without our knowledge or control. To provide confidentiality and privacy is very important today to enterprises and other users to use cloud services. Cloud is becoming a dominant computing platform. Researchers have demonstrated that the essential issue of DDoS attack and defense is resource competition between defenders and attackers and maintain data venerability in the cloud. Major problem in clouds are load balancing and sharing the data to the particular user. In our project will propose a job scheduling and attribute base data sharing.
Published by: Vilas R. Bodkhe
Author: Vilas R. Bodkhe
Paper ID: V4I2-2188
Paper Status: published
Published: May 1, 2018
Geopolymer concrete with replacement of silica fume and foundry sand
A concrete use around the world is second only to water. The production of ordinary Portland cement contributes 5-7% of total greenhouse gas emission. It also consumes large amount energy. Hence it is essential to find the alternative to cement. Silica fume it is an ultrafine powder collected as a by-product of the silicon and ferrosilicon alloy production. It is also rich in silica and alumina. In this paper, silica fume is used to produce a geopolymer concrete. Geopolymer is a material resulting from the reaction of a source material that is rich in silica and alumina with the alkaline solution. Geopolymer concrete is totally cemented free concrete. In geopolymer, silica fume act as the binder and alkaline solution act as an activator. Silica fume and foundry sand are totally replaced and alkaline activator undergoes polymerization process to produce aluminosilicate gel. Alkaline solution used for present study is the combination of sodium hydroxide (NaOH) and sodium silicate (Na2Sio3) with ratio 3. A grade chosen for the investigation were M25.The mix was designed for molarity of 8M. Hot air curing is done by placing the specimens in the electric oven at 60ºC for the 24hours duration and 7days 14 days and 28days test is carried after heating. The geopolymer concrete specimens were tested for their compressive strength, flexure strength and split tensile test. Experimental investigations have been carried out on workability, the various mechanical properties of GPCs.
Published by: Aysha Banu. M, Yuvaraj. R, Saranya. P, Girija. S
Author: Aysha Banu. M
Paper ID: V4I2-2186
Paper Status: published
Published: May 1, 2018
Credit card fraud detection using Naïve Bayes model based and KNN classifier
Machine Learning is the technology, in which algorithms which are capable of learning from previous cases and past experiences are designed. It is implemented using various algorithms which reiterate over the same data repeatedly to analyze the pattern of data. The techniques of data mining are no far behind and are widely used to extract data from large databases to discover some patterns making decisions. This paper presents the Naïve Bayes improved K-Nearest Neighbor method (NBKNN) for Fraud Detection of Credit Card. Experimental results illustrate that both classifiers work differently for the same dataset. The purpose is to enhance the accuracy and enhance the flexibility of the algorithm.
Published by: Sai Kiran, Jyoti Guru, Rishabh Kumar, Naveen Kumar, Deepak Katariya, Maheshwar Sharma
Author: Sai Kiran
Paper ID: V4I3-1165
Paper Status: published
Published: May 1, 2018
Chatbot for education system
The purpose of this paper is to develop an automated system which gives a reply to a user query on behalf of a human for the education system. It can give an answer to each and every query asked by the end user. Existing chatbots such as facebook chatbot, WeChat, Natasha from Hike, Operator, etc. were giving reply from its local database. But our approach is to focus on the local database as well as web database and also to make system scalable, user-friendly, highly interactive. Various techniques such as machine learning, NLP, pattern matching, data processing algorithms are used in this paper to enhance the performance of the system.
Published by: Guruswami Hiremath, Aishwarya Hajare, Priyanka Bhosale, Rasika Nanaware, Dr. K. S. Wagh
Author: Guruswami Hiremath
Paper ID: V4I3-1155
Paper Status: published
Published: May 1, 2018
