Abstract:

One of the significant issues facing web users is the amount of noise in web data which hinders the process of finding useful information in relation to their dynamic interests. Current research works consider noise as any data that does not form part of the main web page and propose noise web data reduction tools which mainly focus on eliminating noise in relation to the content and layout of web data. This paper argues that not all data that form part of the main web page is of a user interest and not all noise data is actually noise to a given user. Therefore, learning of noise web data allocated to the user requests ensures not only reduction of noisiness level in a web user profile, but also a decrease in the loss of useful information hence improves the quality of a web user profile. Noise Web Data Learning (NWDL) tool/algorithm capable of learning noise web data in web user profile is proposed. The proposed work considers elimination of noise data in relation to dynamic user interest. In order to validate the performance of the proposed work, an experimental design setup is presented. The results obtained are compared with the current algorithms applied in noise web data reduction process. The experimental results show that the proposed work considers the dynamic change of user interest prior to elimination of noise data. The proposed work contributes towards improving the quality of a web user profile by reducing the amount of useful information eliminated as noise.

 INTRODUCTION :

NOWADAYS the web is widely used in every aspect of day to day life, a daily use of web means that users are searching for useful information [1]-[3]. However, ensuring useful information is available to a specific user has become a challenging issue due to the amount of noise data present on the web [4]. Noise in web data is defined as any data that is not part of the main content of a web page [5], [6]. For example, advertisements banners, graphics, web page links from external web sites etc. Noise web data elimination is a concept which involves detection of web data that needs to be eliminated because it either does not form part of the main web page content or is not useful to a given user [7]. It is recognised in the current research work [8] that the noise web data reduction process is site-specific, i.e. it involves removal of external web pages that do not form part of the main web page content. However, this work does not focus on the structure and layout of web data to identify and eliminate noise but instead, a key focus is on extracted web log data that defines a web user profile. In view of this research, noise isnot necessarily advertisements from external web pages, duplicate links and dead URLs or any data that does not form a part of the main content of a web page, but also useful information that does not reflect dynamic changes in user interest Various machine learning tools/algorithms are used to discover useful information from web data, this process is referred to as web usage/data mining process [1], [2]. It finds user interest patterns from web log data. Web log data contains a list of actions that have occurred on the web based on a user [9]. These log files give an idea about what a user is interested in available web data. Web log data contain basic information such as IP address, user visit duration and visiting path, web page visited by the user, time spent on each web page visit etc. In this work, web log file and web data are used interchangeably because a log file contains web data, therefore elimination of noise web data is based on extracted web user log file. In a real world, it is practically impossible to extract web log data and create a web user profile free from noise data. A web user profile is defined as a description of user interests, characteristics, and preferences on a given website [10]-[12]. User interests can be implicit or explicit [13]. Explicit interests are where a user tell the system what his/her interests are and what they think about available web data while implicit interest is where the system automatically finds interests of a user through various means such as time and frequency of web page visits [14], [15]. Many users may not be willing to tell the system what their true intentions are on available web In a real world, it is practically impossible to extract web log data and create a web user profile free from noise data. A web user profile is defined as a description of user interests, characteristics, and preferences on a given website [10]-[12]. User interests can be implicit or explicit [13]. Explicit interests are where a user tell the system what his/her interests are and what they think about available web data while implicit interest is where the system automatically finds interests of a user through various means such as time and frequency of web page visits [14], [15]. Many users may not be willing to tell the system what their true intentions are on available web data, therefore, this work will focus on implicit user interests. Current research efforts in noise web data reduction have worked with the assumption that the web data is static [16]. For example, [17], [18] proposed a mechanism where noise detected from web pages is matched by stored noise data for classification and subsequent elimination. Therefore, it shows that elimination of noise in web data is based on pre-existing noise data patterns. In evolving web data, existing noise data patterns used to identify and eliminate noise from web data may become out of date. For this reason, the dynamic aspects of user interest have recently become important [19], [20]. Moreover, web access patterns are dynamic not only due to evolving web data but also due to changes in user interests [21]. For example, web users are likely to be interested in data derived from events such as Weddings, Christmas, Birthdays etc. Therefore, it is necessary to discover where such dynamic tendencies impact the process of eliminating noise from web data. To address dynamic issues in noise web data reduction, this research proposes a machine learning algorithm capable of learning noise in web data prior to elimination. The proposed algorithm considers the dynamic change in user interests and evolving web data to identify and learn noise data. The main novelties of this research are: · To demonstrate how dynamic user interests and evolving web data impact noise web data reduction process. This takes into account contribution made by current research works and their limitations in relation to the current state of the art. · To propose a machine learning algorithm capable of learning noise in a web user profile prior to elimination. Elimination of noise from a web user profile does not only depend on pre-existing noise data patterns, but it learns noise levels based on dynamic changes in user interest as well as evolving web data. · The outcome of the practical application of the proposed tool will reduce the amount of useful information eliminated as noise from a web user profile. This may significantly improve the quality of a web user profile. The rest of this paper is organised into the following sections; Section II positions the proposed work based on current research work. Section III discusses the proposed NWDL process. Section IV is experimental results and analysis. Finally, section V is the conclusion of this paper.

SYSTEM REQUIREMENTS
SOFTWARE REQUIREMENTS:
• Programming Language : Python
• Font End Technologies : TKInter/Web(HTML,CSS,JS)
• IDE : Jupyter/Spyder/VS Code
• Operating System : Windows 08/10

HARDWARE REQUIREMENTS:

 Processor : Core I3
 RAM Capacity : 2 GB
 Hard Disk : 250 GB
 Monitor : 15″ Color
 Mouse : 2 or 3 Button Mouse
 Key Board : Windows 08/10