presented by:
B.DIVYA
N.L.MANASA

---

[attachment=11600]
Objective
This project can suggest and rank the popular item based on the user selection. This ranking process can be achieved by using the naïve algorithm.It has 2 rules.
1.ranking rule and
2.suggesting rule.
Existing System
Suggested data’s are not the best user required data. This system does not provide the information related to that user query.
Feedback is established through suggestions of different users. This feedback is not sufficient to ranking the item.
Selection of items is not restricted to items presented to the user.
Graphical representation of Result is not possible.
Proposed System
In this system Lightweight and self tuning algorithm is used. This can provide the information related to the user preferred.
This system can rank the item depends upon the feedback of the each and every user feedback.
No configuration parameter is needed to get the better result.
Software Requirements
Operating system :- Windows XP Professional
Front End :- Visual Studio 2008, ASP. net, C#
Backend :- SQL Server 2005
Hardware Requirements
SYSTEM : Pentium IV
HARD DISK : 40 GB
RAM : 512 MB
Modules
Login
Registration
Admin
Search products
Rank the Popular Item
Buy products
Module Description
Login:
User has to give their username and password. If the username and password is valid then that user can be permit to access this web application.
This module will allow only registered users to access.
Registration:
New user can’t search the product directly. They want to register here to use this application.
User wants to provide all the required details in registration form.
Registered user name and password is considered as valid.
Admin:
In this module, Admin have rights to upload as well as update product details in the application.
Admin can delete the product if it is not required.
Algorithm
A Naive Algorithm:
In this project naïve algorithm is used to rank the popular items based on number of selection of this item for the past. In this algorithm whenever the item is selected, item count will be incremented by one. Based on the item count item will be ranked and suggests the users about the popular items to buy. So that users can consume their time.
This project can have some rules they are, ranking rule and suggesting rule
TOP (Top popular) Init:
Vi = 0 for each item i
At the t-th user selection:
If item i selected:
Vi ? Vi + 1
S ? a set of s items with largest V countsM2S (Move-to-set)
Ranking rule: The rank scores are updated only for an item that was selected and was not suggested to the user. This can increase count of the non suggested item, and also rank the item based on the selected items
Suggesting rule:
It suggests the user based on the previously selected item by the user or customer. This rule can suggests the item by recently chooses item by counting them.
At the t-th user selection with item i selected:
If item i not in the suggestion set S
Remove a random item from S
Add i to S
Ranking rule2 ::
Init: Ti = 0, Vi = 0, for each item I
At the t-th user selection:
For each item i:; If item i not suggested:
Ti ? Ti + 1 If item i selected:
Vi ? Vi + 1;?i ? Vi/Ti
Data Flow Diagram
USECASE DIAGRAM
Advantages
This system can suggest the users while searching the product in application.
It can allow the user to search their preferred product.
This can rank the item automatically based on their user selection.
This application can list the best product to the user default.
User can get the product from that list as they want.
Applications
Authorized user can login into this application.
This system can have admin control. In this admin can upload the details of the product.
User want select their preferred category to search the product.
Based on the user category the item can be listed in the display.
The item can be ranked based on the frequently selected items from the list.
This system can suggest the user to search the category.
Conclusion
This system can use the simple randomized algorithm for ranking and suggesting the popular item. This can suggest the user to search the product. It can rank the popular item based on the previous user selection. This system can give the preference to non suggested product also. This system can be suggest user and rank the popular item by using the proposed algorithm. Focused on understanding the limit ranking of the items provided by the algorithms, and how it elates to that of the true popularity ranking and assessed the quality of suggestions as measured by the true popularity of suggested items.