Abstract: Web Mining is playing a major role in Networks. This paper reflects the project work of two important problems identified during master degree. The first case study is about spamming. E-mail spam. Spammed content may contain many copies of the same message. In existing work, various filtering techniques are used to detect these e-mails such as Random Forest, Naive Bayesian, Support Vector Machine (SVM) and Neutral Network. In this project work Naive Bayes algorithm is chosen for e-mail spam filtering. Two datasets Spam Data and SPAMBASE datasets was selected for this project work. The execution of the datasets is evaluated based on their accuracy, recall, precision and F-measure. This work use VB.Net and C# for the implementation of Naive Bayes algorithm for e-mail spam filtering on both datasets. The outcome shows that the type of email and the number of instances of the dataset has an influence towards the performance of Naive Bayes.
The second case study is about finding optimization solution for shortest path. MLT (Machine Learning Techniques) finds potentially useful patterns in the data. The famous example is probably the Traveling Salesman Problem (TSP) in which a salesperson intends to visit a number of cities exactly once, and returning to its starting point, while minimizing the total distance traveled or the overall cost of the trip. TSP is used in combinatorial optimization. This paper proposed genetic algorithm based cuckoo search technique for TSP. Result significantly improves the performance of the TSP. The algorithm is implemented using MATLAB.
Keywords: Naive bayes algorithm, Genetic algorithm, Cuckoo search.
| DOI: 10.17148/IJARCCE.2020.9531