Problems of current system:
After reviewing several nutrition systems, we summarized four main deficiencies among these systems. The following is the detail description of each deficiency.
Without server support
Nearly all the processes of systems that we studied are operated on client side, which required device with vital computing power and storage space. Unfortunately most PDA’ CPU is around 200Hz to 300Hz throughout the market. Therefore reserving large calculation and database on the PDA may slow down the operation. That is why our proposed system is designed as server-client architecture. We only keep some general database and simple calculation on client side. The server side will execute most tasks like large graph presentation, calculation, or query. From another point of view, our system can be considered as just an input device.
Without motivation for data entering
One common problem of there systems is that there is no back up method when users don’t enter the diet information sometimes. If this situation happens all the time, the figure would not be consisted with user’s real condition. In order to encourage people to enter data frequently and regularly, we offer some promotion plan after their registration. If user really can’t enter, we will estimate the users’ situation according to their diet style and health information they provided.
Without notification support
The third problem is that most nutrition softwares can not provide notification for user under some situations. Take a person with heart disease as example. When he wants to order a dish that is not good for his health at a restaurant, at this moment the system will warn him to choose another one or give him some suggestion. Or when he is confusing which dish is better, system will compare the nutrition fact of those dishes so as to make a recommendation for user. As we know that people with heart disease have to take pill regularly, therefore our proposed system have the function that alert user to take pill after he set up the schedule on the system.
Inefficient database update
The last problem of these systems is inefficient database update, especially on food selection. For most current systems, the database of food only consists of general food. If the user eats some diet that is not on the list, users can enter the food by themselves. Although most system support this function, it would be not so efficient when number of the data to be entered is large. Our proposed system provides the function that users can update their database from server whenever the users is online.