Medical study has revealed that people set a bigger possibility of countering free radicals and warding off illness by consumption of healthy foods and by increasing their resistant system. Due to the poor eating habits people suffer from many diseases. In the current scenario fast food become important food in daily routine because it is effortlessly available but taking fast food in routine may cause for disease like heart attack, diabetics etc. Healthier diets help us to maintain our health and keep us away from many diseases. For better recovery from diseases or surgery etc individual have special needs according to their medical profile, cultural backgrounds and nutrient requirements. Design and implementation of healthy diet recommendation system is based on web data mining which is the application of data mining technique help us to determine pattern from web. In terms of accuracy and time performance analysis of recommendation system using two decision tree learning algorithm ID3 and C4.5 and apply it on healthy diet application. In a machine learning process the classification can be de-scribed as a supervised learning algorithm. Data records are belong to class on the bases of knowledge of class it assign a class labels to data to co- design and co develop software and hardware, and hence, such components. However, incorpora-tion of that deal with knowledge extraction from database records and prediction of class label from unknown data set of records. We can define classification is a development in which specified set of data records is separated into training and test data sets. For validating the model we required the test data record and for constructing the classification model training data set is required. The constructed classification model is used for classifying and predicting new data set records. These new data set records are different from training and test data set. For getting higher classification accuracy or accurate prediction we required a prior knowledge of the class label data record which makes attribute selection effortless. For higher classification accuracy supervised learning algo-rithm (like classification) is preferred to unsupervised learning algorithm (like clustering). In current scenario, data mining technology has been widely used in education, real estate, stocks, health care and other fields. A number of widespread classification algorithms used in data mining and decision support systems is: neural networks, logistic regression, Deci-sion trees etc. Among these classification algorithms decisiont-ree algorithms is the most frequently used because of it is ef-fortless to understand and cheap to implement the article is being submitted to and the manuscript identifi-cation number. Click the forward arrow in the pop-up tool bar to mod. In the data collection and preprocessing web server data base contains two types of data base one is content data base that contain the information like user information and other types of data and second is the server log data base for recording the HTTP transaction (log records).Data collection or data acquisi-tion module collect data from the external web atmosphere to provide resources and material for the latter data mining. From the web environment the data source we get the web. Data preprocessing mainly renovate and progression the source data acquired in data collection phase and construct the data warehouse of associated themes to generate basic plat-form for data mining process. Data preprocessing is prepara-tion for data mining and it mainly includes data scrubbing, data integration, data conversion, data reduction, etc...Basically in the data preprocessing step convert the data into the form which is accepted by the data mining algorithm.