The interactions with web systems is huge because of COVID-19, where every system is run through online interactions. The world wide web is continuously expanding, and users' interactions with websites generate a vast quantity of data. Web usage mining is the use of data mining techniques to extract important and hidden information about users. It allows you to see the most frequently visited sites, imagine user navigation, and track the progress of your website's structure, among other things. The web mining techniques help us to analyze the user's behavior and accordingly create the required web designs, which will appear in the relevant searches of the users. In this scenario, one of the important processes is web document preprocessing, which will help us to extract the particular quality data inputs for analyzing the behaviors which helps in effective web design. Here, the authors discuss preprocessing of web documents. From the four different phases of the web mining, web document pre-processing is a very important phase.
TopWeb Mining
The World Wide Web is a collection of web sites that provides internet users with a wealth of information. For internet users, the knowledge available on the internet has evolved into a valuable resource. Because the number of websites available on the internet is growing and becoming more complicated, the total volume of web is enormous. A website serves as a link between the customer and the business. The corporations can update visitor’s performance during web inquiry, and identify the trends. Web mining is defined as the search for and analysis of useful information on the World Wide Web. Web content mining, web structure mining, and web use mining are the three types of web mining. The extraction of useful information and online knowledge from web resources or web contents such as text, picture, audio, video, and structured data is referred to as web content mining (Mehra & Thakur, 2018). Web use mining may be defined as the discovery and analysis of user access patterns using log file mining. The WUM's output may be utilised for web personalisation, recovering system performance, site modification, and use description, among other things. Web log file is a server log file that contains access logs of the web server and is a vital data source in Web use mining. The Data Preprocessing segment is a crucial stage in the WUM. Data cleansing, session identification, user identification, and path completion are all included. Material preprocessing is used to remove unwanted data from log files so that the pattern discovery algorithm can detect the user pattern (Anand & Aggarwal, 2012).
Need of Web Mining
Web mining is the use of Data Mining methods to locate and extract information from Web publications and services automatically. Web mining's major goal is to extract relevant information from the World Wide Web and its usage trends. The focus on web mining in academics, the software business, and online-based organizations has resulted in a substantial amount of expertise. By recognizing online pages and categorizing web content, web mining helps to increase the power of web search engines. E-commerce websites and e-services benefit greatly from web mining (Chu et al., n.d.).
Types of Web Mining
As seen in Figure 1, web mining may be separated into three types.
The practice of obtaining meaningful information from the content of Web pages is known as web content mining. The contents of a web document relate to the notions that the page was designed to convey to users. Text, picture, video, music, and records like lists and tables can all be used to create this content. Text mining has received greater attention than other fields (Jokar et al., 2016).
The web may be seen as a graph, with nodes and edges representing the connections between documents. The method of obtaining structural information from the web is known as web structure mining (Jokar et al., 2016).
Web use mining is the use of data mining techniques to uncover trends on the Internet in order to better understand and satisfy the requirements of users. This sort of web mining looks at information about how people utilise the internet. It's worth noting that there aren't any apparent distinctions between web mining groups. Web content mining algorithms, for example, can leverage user information in addition to documents. It is also possible to attain greater outcomes by combining the approaches mentioned above (Jokar et al., 2016) (Han et al., 2011).