ABSTRAT:
The machine learning algorithm has become important because of their accuracy in forecasting. It is very difficult to predict a customer’s performance due to an unexpected customer situation. Many algorithms are designed for the same purpose. In this paper, we have studied and analysed three Bays algorithms such as AODE, Naive Bayes and AODE sr. We implemented these algorithms in the WEKA tool and built a new model that provides better accuracy than the existing one. During development we have tried to reduce the noise and error in the data and we also need to filter the information. This process will allocate Wj weight to the new filtered data. The error can be demarcated as: E (j, k), Where j ᒈ J or it’s an assumption. k is a function of purpose. Similarly noise can be defined by another function N = E + Wj .
EXISTING SYSTEM :
In this type of learning, information is available in advance. In order to ensure adequate allocation of data to groups of algorithms, that should be explained. In other words, the system learns on the basis of input and output power. In supervised learning, the program manager, who acts as a type of teacher, gives the correct amount of feedback. The purpose is to train the method in the perspective of sequential input and output calculations and establish communication. The Naive Bayes is a model of probabilistic distinctions, based on the concept of autonomy. Though, in numerous real-world mining applications, this statement is often dishonoured. In response to this statement, scholars have done a great deal of testing the correctness of NB by abating the quality of their stability
EXISTING SYSTEM DISADVANTAGES:
1.LESS ACCURACY
2. LOW EFFICIENCY
PROPOSED SYSTEM :
have proposed an idea named Averaged One- Dependence Estimators (AODE) that decreases the independent predictive value by sampling all prototypes from a constrained class of dependent classifiers. Inspired by this research, we rely on that passing on diverse value to these different classifiers can lead to greater enhancement. We have experimentally verified our algorithm with Weka tool[2], using Super Market data sets and briefly defined a comparative study between Naive Bayes, AODE and AODE sr. The investigational outcomes indicate that proposed algorithm meaningfully leave behind all the other algorithms used to compare
PROPOSED SYSTEM ADVANTAGES:
1.HIGH ACCURACY
2.HIGH EFFICIENCY
SYSTEM REQUIREMENTS
SOFTWARE REQUIREMENTS:
• Programming Language : Python
• Font End Technologies : TKInter/Web(HTML,CSS,JS)
• IDE : Jupyter/Spyder/VS Code
• Operating System : Windows 08/10
HARDWARE REQUIREMENTS:
Processor : Core I3
RAM Capacity : 2 GB
Hard Disk : 250 GB
Monitor : 15″ Color
Mouse : 2 or 3 Button Mouse
Key Board : Windows 08/10
