Authored on: July 15, 2011 by Jeffery Dilegge
Data analysis, also known as data modeling is the process of collecting, inspecting and filtering data so that certain sorts of data can be highlighted for closer inspection. Data analysis has many purposes and techniques. This is just going to focus on a couple of popular topics, later I will get into more detail with data analysis.
Data mining is the process of collecting data based on specific patterns in the target environment. Data mining is very important to the scientific process and to business research, however in this tutorial, I will be breaking it down to spatial data mining. Data mining is broken down into 4 classes of steps which are associational rule learning, clustering, classification and regression. There are a few types of data mining, we will be looking mostly at Pattern Mining, which is data mining via the associative patterns of a data set.
Associational rule learning is simply the focus on locating common patterns in data substrings that will allow for the easiest but most accurate collection and analysis of data. For example, in PHP I would use preg_match() to target specific portions of a string while ignoring all other data. Without the associational process, I would have no substring pattern to search for and would be shooting in the dark. In PHP, you could use something like this to locate the description meta tag within a webpage.
<?php
preg_match("|<meta[^>]*description[^>]*content=\"([^>]+)\"[^>]*>|Ui",$string, $results);
?>
Clustering is what the process is called when one locates and discovers groups and structures that are similar in some way to help extend the associational rule data. Usually, this process is used for more thorough and complex data mining that is using many filters to extract a larger number of substrings from the target subject.
Classification Classification is just what it sounds like, the labelling of data groups based on their distinct properties that separate said substring or data group from other data groups.
Regression The process in which one would create or locate a tool or function to model or process the data that was collected from the previous steps.
Packet Sniffing, also known as network analysis, record and analyze data transferred between two network connected devices for the purpose of troubleshooting network issues, gathering network usage statistics, security intrusion detection, statistic gathering and for spying on network users. Network analyzers come in software and hardware versions. Software based wifi network analyzers run on of the devices being monitored, where as hardware used for packet sniffing typically work just like a router. Software network packet sniffers are usually used for monitoring server traffic and the invasion of privacy or others. Hardware network packet sniffers are most commonly used for diagnostics, security related purposes and monitoring for performance.
Packet sniffing, when used for blackhat purposes like the invasion of privacy, is extremely illegal and governed by multiple Federal laws in the United States. The Wiretap Act states, "to intercept and monitor communications placed over their facilities in order to combat fraud and theft of service." Thus, packet sniffing someone's wireless connection is illegal, unless permission has been granted by the police or other higher powers in the form of a contract or warrant.
Packet sniffing in the wrong hands allows hackers to spy on Internet user activities for the purpose of collecting private information. This is why it is so very important to use a secured Wifi connection, or on wired connections, to use strong firewalls both software and hardware. Most modern Internet routers have a built in hardware firewall that is on by default but you still need a software firewall on every computer on the network.
|
