• No results found

A SECURE AND PERSONALIZED WEB SEARCH WITH ONTOLOGY

N/A
N/A
Protected

Academic year: 2022

Share "A SECURE AND PERSONALIZED WEB SEARCH WITH ONTOLOGY"

Copied!
5
0
0

Loading.... (view fulltext now)

Full text

(1)

129 | P a g e

A SECURE AND PERSONALIZED WEB SEARCH WITH ONTOLOGY

M. Lakshmi Madhuri

1

, M. Purushotham Reddy

2

, G. RamaSubbaReddy

3

1

M.tech Scholar (CSE),

2

Asst.professor, Dept. of CSE, Vignana Bharathi Institute of Technology (VBIT),Vidya Nagar, Pallvolu, Proddatur, Kadapa (Dist),Andhra Pradesh (India)

3

Working as Associate Professor and Head of Department (CSE),

Vignana Bharathi Institute of Technology (VBIT),Vidya Nagar,Pallvolu, Proddatur, Kadapa(Dist), Andhra Pradesh (India)

ABSTRACT

Personalized web search has displayed that how much effective they are when it comes to improvement of quality for various search services over the internet. There are proofs which show that users are prone for not to hide their private data during the process of search so this has become a main stoppage for the personalized web search. In this paper we will be discussing about privacy of personalized web search applications. So we will give personalized web search framework known as user profile systems which will acquire profiles of the users using queries taking care of user’s privacy requirements. In this paper we will prepare a runtime generalization which aims at making a momentum between two metrics which executes privacy of personalized data. In this paper we propose two greedy algorithms, named GreedyDP and GreedyIL, for runtime generalization. Online prediction mechanism will be provided for deciding whether personalizing a query is beneficial or not. Broad experiments display the effectiveness of framework.

I. INTRODUCTION

Nowadays web search engine has become a tool for many users using web for obtaining useful information on the web. Sometimes users may face failures when search engine returns unmatched results which do not matches real intentions. These unmatched results are mainly due to large number of users as well as duplication of texts. Customized web look is a general classification of inquiry procedures going for giving better indexed lists, which are custom-made for individual client needs. As the cost, client data must be gathered and examined to make sense of the client aim behind the issued question. The answers for personalized web search can for the most part be arranged into two sorts, in particular snap log-based strategies and profile-based ones. The snap log based systems are clear they basically force inclination to clicked pages in the client's inquiry history. Despite the fact that this method has been shown to perform reliably and extensively well, it can just chip away at rehashed questions from the same client, which is an in number impediment restricting its materialness. So we can say that privacy concern has become important barrier for personalized web search services. Now we will be discussing about personalized web search which already exists so system which already exists does not have any support for runtime profiling. This means that user profile is generally used only once when it is offline which results in personalizing all the queries from same user unwontedly. So by personalizing all queries for same user will ultimately results in data privacy risk of the user because it has been exposed to the server.

(2)

130 | P a g e Therefore to deal with this problem we will be proposing a system for preserving privacy of the user by giving a framework of user profile system which will be used for personalizing profiles for every query according to user privacy requirements. In this paper we propose two greedy algorithms, named GreedyDP and GreedyIL, for runtime generalization. Online prediction mechanism will be provided for deciding whether personalizing a query is beneficial or not. Broad experiments display the effectiveness of frameworks.

II. RELATED WORK

Personalized web search came into existence many years ago during those times many personalization strategies has been interrogated, still work is going on whether personalization is really consistent and effective on various queries of several users and also on several search contexts. In this paper we will be analyzing this problem and give some valid conclusions. There are various ways to provide personalized web search results one of the most approved way is by providing user profiles and descriptions of user’s point of interest which also can be used by search engines. There are many ways for creating user profiles to collect user information with the medium of proxy servers or through desktop bots. But these ways or approaches are possible when there is user involvement in it for installing proxy servers and bot. In this paper we will be also working on retrieval process which means whatever user has searched up till now in various web search engines we will be trying to get back search history based on user’s search preferences. There are already some retrieval processes which exist but they are having some deficiency in it means they are lacking user modeling and also they are not prone to single user which ultimately results in low retrieval performance. In this paper we will be providing low cost mechanisms for the client to personalize the query for the client in user profile system.

III. SYSTEM DESIGN

In this we will be discussing about design of the project so first thing is data flow diagram which is also called as bubble chart. Data flow diagram is a simple graphical representation of data which is used to represent a system in terms of input data to the system; data is undergone through various numbers of processing and according to that output data is generated by the system.

Data Flow Diagram (User)

(3)

131 | P a g e Component Diagram (User)

A greedy algorithm is a mathematical process of creating a set of attributes from smallest parts. We will be following recursion concept for solving problem based on the solution of that problem and again based on solution of smaller part of the same problem. Generally recursion is the concept of calling function within itself that’s why greedy algorithm will be used for calling the problem within itself so we will be using greedy approach for protecting privacy of data because it is simple, with this we can implement very complex solutions easily and also it will deal with multi step problems so that we can know next step and also the benefit of it.

Greedy algorithm is called as greedy because it will solve the problem and its attributes by creating objects of the problem which means smaller parts of the problem simultaneously providing immediate outputs. Greedy algorithm does not take larger problem as a one thing but it will first decide the solution and then execution once decision has been made it is never reconsidered. So we can say that advantage for greedy algorithm will be that it will create smaller objects of the large problems which will be straight and easy to understand. After every advantage their will one disadvantage so for greedy algorithm disadvantage will be that once it creates smaller solutions of large problems it will lead to long lasting wrong outcomes. Now we will discuss uses of greedy algorithm it is used in mobile networking for providing route packets with less number of hop count and shortest delay also. Greedy algorithms are used in machine learning process, business intelligence, and artificial intelligence.

IV. MODULES DESCRIPTION

In this paper we will use four modules which are as follows-:

1). Profile Based Personalization 2). Generalizing User Profile 3). Online Decision

4). Privacy protection in personalized web search system.

Profile based personalization means that we can edit the profile of the users by using greedy approach, generalizing user profile means we can edit user profile by using greedy approach, online decision means that we can make decision online for greedy algorithm which means that greedy will make decision so in that case we will be using online decision module, and last is privacy protection in personalized web search system which

(4)

132 | P a g e means that user’s data protection is the main thing so we will propose a system for protecting user data and to make it risk free also.

V. OBJECTIVES

Now we will be discussing some objectives of the personalized web search system which are listed below-:

A). Process of converting user description as an input into a computer based system is called as input design.

This design is used for avoiding errors in the user’s data which is as an input process and to show rite path to the management for retrieving correct information from the computer based system.

B). above objective can be obtained by forming user friendly screens for data entry purpose so that it can handle large amount of data. So the main aim of input design system is to make entry of data much easier and also to make it bug free. If we need any data manipulation we can achieve it by designing screen in such a manner and also it provides facility for viewing recording.

C). Third and last objective will be that when we will enter data first thing it will perform is validity checking.

As we know that data will be entered in a screen i.e. large screens. So according to those proper messages will be provided so that user will not be in perplexed situation. So we can conclude by saying that input design is used to create input layout which is easy to look for.

VI. RESULT

In this paper our aim is to provide a quality output not only quality outputs but also check whether it meets requirements of the end user and whether it is providing clear information or not. Usually in other systems also outcomes are processed and communicated to other users as an output. It should display output in the form of hard copy also which means that if anyone need information of the user then one print out should be ready for immediate action. So it will benefit the user when they need direct information that is the reason we will perform at last hard copy for an output. If the output design is according to the expectations then it will help user to establish a system relationship which will ultimately help user for decision making only if the output design is efficient and intelligent.

VII. CONCLUSION

At last we conclude by saying that personalized web search has a framework known as user profile system which will be used for personalizing profiles for every query according to user privacy requirements. In this paper we propose two greedy algorithms, named GreedyDP and GreedyIL, for runtime generalization. Online prediction mechanism will be provided for deciding whether personalizing a query is beneficial or not. Broad experiments display the effectiveness of frameworks. There are many ways for creating user profiles to collect user information with the medium of proxy servers or through desktop bots. But these ways or approaches are possible when there is user involvement in it for installing proxy servers and bot.

(5)

133 | P a g e

REFERENCES

[1] Z. Dou, R. Song, and J.-R. Wen, “A Large-Scale Evaluation and Analysis of Personalized Search Strategies,” Proc. Int’l Conf. World Wide Web (WWW), pp. 581-590, 2007.

[2] J. Teevan, S.T. Dumais, and E. Horvitz, “Personalizing Search via Automated Analysis of Interests and Activities,” Proc. 28th Ann. Int’l ACM SIGIR Conf. Research and Development in Information Retrieval (SIGIR), pp. 449-456, 2005.

[3] M. Spertta and S. Gach, “Personalizing Search Based on User Search Histories,” Proc. IEEE/WIC/ACM Int’l Conf. Web Intelligence (WI), 2005.

[4] B. Tan, X. Shen, and C. Zhai, “Mining Long-Term Search History to Improve Search Accuracy,” Proc.

ACM SIGKDD Int’l Conf. Knowledge Discovery and Data Mining (KDD), 2006.

[5] K. Sugiyama, K. Hatano, and M. Yoshikawa, “Adaptive Web Search Based on User Profile Constructed without any Effort from Users,” Proc. 13th Int’l Conf. World Wide Web (WWW), 2004.

[6] X. Shen, B. Tan, and C. Zhai, “Implicit User Modeling for Personalized Search,” Proc. 14th ACM Int’l Conf. Information and Knowledge Management (CIKM), 2005.

[7] X. Shen, B. Tan, and C. Zhai, “Context-Sensitive Information Retrieval Using Implicit Feedback,” Proc.

28th Ann. Int’l ACM SIGIR Conf. Research and Development Information Retrieval (SIGIR), 2005.

[8] F. Qiu and J. Cho, “Automatic Identification of User Interest for Personalized Search,” Proc. 15th Int’l Conf. World Wide Web (WWW), pp. 727-736, 2006.

AUTHOR DETAILS

M. Lakshmi Madhuripursuing M.Tech (CSE) Vignana Bharathi Institute of Technology (VBIT),VidyaNagar, Pallvolu, Proddatur, Kadapa (dist),Andhra Pradesh 516 362

M. Purushotham Reddy received his M.Tech (Computer Science & Engineering) from

Jawaharlal Nehru Technology University, Anantapuramu and pursuing Ph.D in JNTUA , Anantapuramu. Presently he is working as Associate Professor in Computer Science &

Engineering, Vignana Bharathi Institute of Technology, Proddatur, Kadapa dist, A. P., India.

G.RamaSubbaReddy received his M.E (Computer Science &Engineering) from Sathyabama

University, Chennai.Presently he is working as Associate Professor and Head of the Department in Computer Science & Engineering, Vignana Bharathi Institute of Technology, Proddatur, Kadapa Dist.,A.P, INDIA

References

Related documents

JSC ARMZ Uranium Holding Company (as it is now known) became the mining division of Rosatom in 2008, responsible for all Russian uranium mine assets and also Russian

The user experiences and is affected by transitions between the layouts (e.g., webpages or mobile app screens) of interactive systems. Such transitions affect the

iii) Once a new CRN is selected, click the link for Midterm Grading from the Advisors Menu. i) Only unsatisfactory grades are required, however you may enter all grades if you

The proposed methodology has a number of novel features: (i) both the mean function and the covariance function can be estimated simultaneously under a set of mild

The small cell size phenotype of tsc1 or tsc2 cells is reversed by tor1 disruption: While both Tor1 and Tsc1/2 positively regulate amino acid uptake, cell size appears to be

This study analyzed three 19 th century North Carolina history textbooks to determine if there were any changes in the historical narratives presented in these textbooks over the

whether the above observed rad9 mutant phenotypes are unique or common to other G 1 /S and G 2 /M check- points, we combined a rev3 mutation with additional checkpoint mutations,

It emerged that the examined economies generally offer a range of support measures, including working capital augmenting facilities, pre-shipment financing, and