الفهرس 
ContentsOnly 14 pages are availabe for public view
 |  | from | 144 |  |  |
| | | from | 144 | | |
Abstract
With the spread of e-commerce fields over the world especially an online auctions segment, the problems of internet crimes increased in the last years. The online auctions face a serious problem of trust among the participants where users have no information about the others, and have no relations among them except the commercial transactions; this allows fraud to occur by malicious users. Many trust model have been introduced to measure trust or avoid fraud but all of these models has failed in introducing a simultaneously, accurate, and intelligent mechanism for trust measurement behavior The rest of this Thesis is organized as follows:Chapter 2 is dedicated to related work.Chapter 3: Introduces a new Intelligent Learning for Trust Measurement Behavior (ILTMB) framework.Chapter 4: Buyer Analyzer Layer (BAL) defines the winner of the current transaction bid. Chapter 5: New user (Seller/Buyer) analyzer layer (NUAL): verifies and authenticates all steps of any users. Chapter 6: An Artificial Intelligence Single Item Model Chapter 7: Finally, Conclusions and future work are given. Chapter 1 Introduction: online auctions are increasingly becoming popular, there are many types of auctions such as an English or Dutch. Most of the online auction has many benefits has been discussed Finally, we can see the main challenges of online auction how to measure trust and predict fraud. It clear now the common problem in internet crimes is fraud. Chapter 2 related work: The fraud during e-auction processes reflects bad reputation for this auction and its participants’ so thus reduce the demand for these online auctions. Many studies discussed how to protect the participants from fraud using many techniques. one uses trust models to calculate each participant trust, other one uses special comparable algorithms to predict that the auction users are trusted or un trusted, some auctions may use a third party like payment services as escrow services. collected data and feedback comment from auction users are classified and processed using different algorithms to measure trust Chapter 3 A Proposed Trust Model for Online Auction: This chapter introduces a new efficient framework for accuracy behavior trust measurement, in this framework Intelligent Learning Trust Measurement Behavior (ILTMB) is used not only to predict the trust of the users but also to avoid all types of fraud depending on recommendation system. The recommendation is analyzed through domains classifier which have good impact in the overall system performance. The proposed framework (ILTMB) consists of three layers which are: (i) Historical Domain Analyzer layer (HDAL) (ii) Buyer Analyzer Layer (BAL) (iii) New User (Seller/Buyer) Analyzer Layer (NUAL) All data of users’ profiles will be collected and continuously updated and stored in a single database that has three tables which is called user profile (UP).. HDAL consists of two models filter module (FM) hybrid binary -class and HCM hybrid class module, BAL layer is designed to handle buyer transaction via bid increase price and displayed recommendation list to help the buyers and NUAL layer this layer is designed to register a new seller or buyer. Chapter 4 Buyer Analyzer Layer: The second layer of The proposed framework (ILTMB) will be described. A linear regression model is proposed for analyzing bid behaviors and calculate apart of the trust Chapter 5 New User (Seller/Buyer) Analyzer Layer: Verification and authentication steps very important for any new users: They can be of two types either buyer or seller. Each user, when he/she wants to interact with any e-commerce. Web server for participating in any auction process or to buy goods, in order to prevent online auction fraud, the user should be verified this verification is done through registration steps Chapter 6: An Artificial Intelligence Single Item Model Behavior negotiation, decision making and emotions are the three-main challenge for the single item model so a popular mechanism behavioral tree depending on An Artificial Intelligence has been introduced to enhance the trust equation Chapter 7: Finally, Conclusions and future work are given