Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 23

Transfer to Rank for Top-N

Recommendation
Guided By Group Members
Ms. LEENA Y. • ARATHI KRISHNAN
HEAD OF THE DEPARTMENT • ARCHA RAJAN
DEPT. OF COMPUTER SCIENCE AND • NITIN ULLAS
ENGINEERING
• SEENA PONNACHAN
T.K.M. INSTITUTE OF TECHNOLOGY
ABSTRACT

• Novel and generic transfer learning based recommendation framework coarse-


to-fine transfer to rank is designed.

• Three-staged solution that progressively models users’ preferences from a


coarse granularity to a fine granularity is developed.
• Appropriate recommendation algorithm with pointwise or pairwise
preference assumption is adopted.

• Empirical studies on two large and public datasets showcase merits of


solution in comparison with state-of-the-art methods
INTRODUCTION

• Goal of recommender systems -offer users suggestions about what they might like to
purchase or consume.

• Collaborative filtering aims at modeling users’ preferences relying on the interactions


between users and items.

• Rating prediction and item ranking -modeling directions.

• CF needs to estimate the score of each unrated item for a given user.

• Memory-based CF, model based CF.


• Summary:

• Novel, generic, flexible transfer learning framework to address important top-N


recommendation problem with users’ explicit feedback is proposed.

• Framework is designed so that it is able to model and simulate a user’s shopping


process in a coarse-to-fine manner.

• Extensive empirical studies conducted on two large datasets and find that solution
performs significantly better state-of-the-art methods.
Collaborative filtering
AREA OF IMPLEMENTATION

• Data mining is looking for hidden, valid, and potentially useful patterns in huge data
sets. Data Mining is all about discovering unsuspected/ previously unknown
relationships amongst the data.

• It is a multi-disciplinary skill that uses machine learning, statistics, AI and database


technology.
• The insights derived via Data Mining can be used for marketing, fraud detection,
and scientific discovery, etc.

• Recommendation Site basically are data filtering tools that make use of algorithms
and data to recommend the most relevant items to a particular user. Or in simple
terms, they are nothing but an automated form of a “shop counter guy”. You ask him
for a product.
SYSTEM ARCHITECTURE
System architecture

• The main modules are:-


• Admin
• Constructor
• Event management
• Developer
• Client
MODULES

ADMIN

• Admin has the overall control of the system

• Admin monitors the feedback sent by the user

• Manage Constructors

• Manage event management

• Manage developers

• Manage category

• Monitor rates
CONSTRUCTOR
• View all new construction proposal

• Add bid for each proposal

• Manage all current works

• View rating

• Monitor each clients


DEVELOPER

• View all new developing works


• Add bid for the works
• Manage all current works
• View rating
• Monitor each clients
EVENT MANAGEMENT

• Add different events

• View new events

• Bid for each events

• Manage all current works

• View rating

• Monitor each clients


CLIENTS
• Clients are the normal users or the public
• Add required proposal
• Monitor all bids using KNN algorithm
• Manage bids
• Accept required bids
• Monitor works
• Generate rates
KNN ALGORITHM

1. Load the data

2. Initialize K to your chosen number of neighbours

3. for each example in the data

a) Calculate the distance between the query example and the current example from the

data.

b) Add the distance and the index of the example to an ordered collection.

4. Sort the ordered collection of distances and indices from smallest to largest (in

ascending order) by the distances.


5. Pick the first K entries from the sorted collection
6. Get the labels of the selected K entries
7. If regression, return the mean of the K labels
8. If classification, return the mode of the K labels
KNN ALGORITHM
SYSTEM SPECIFICATION

Hardware Requirements
• System : Pentium IV 2.4 GHz.
• Hard Disk : 40 GB.
• Floppy Drive : 1.44 Mb.
• Monitor : 14’ Colour Monitor.
• Mouse : Optical Mouse.
• Ram : 512 Mb.
Software requirements
• Operating system : Windows 7 Ultimate.
• Front-End : ASP.Net with C#
• Data Base : SQL Server 2008.
CONCLUSION

• From this project we were able to implement the coarse to fine transfer to rank algorithm
to an heterogeneous environment.

• In this project, various needs from the public were read and recommendations were made
in accordance to the need and the capability of the firm to meet the need.

• This perspective is based on a reverse engineering and decomposition of users’ various


needs, i.e., from examination to posting stage to final bidding and selecting the firm.
FUTURE ENHANCEMENT

• In future we can implement this project with another Machine Learning algorithms, which
can improve the accuracy of the model.

• We can implement this project in broader fields.


THANK YOU

You might also like