Assignment 4:
Data Mining
Data mining is consider analyzed data-collect from different aspects and summary that is transformed into useful information. Companies does this
Predicive analytics is a predictor which is measured per customer. For instance, recency, which prefers to the past customer to recent customers.there will be a higher value from past customers then present. "That means that if you contact your customers in order of recency -- first, call the most-recent customer; next, call the next-most-recent customer; and so on -- you will improve your response rate." There are predictors that rank the customer database. Customers with less login time will not likely to renew an annual subscription.
Data mining includes interesting applications. Some are record customers activities of shoppers over a period of time. Web retailers use pattern to better improve items on their website. For instance, telephone companies mine customer who spend more on service than average. They can be pin pointed to buy additional services. This are trends that are used by hospitals, banks, non profit organizations and etc.
"Clustering is a typical unsupervised learning technique for grouping similar data points. A clustering algorithm assigns a large number of data points to a smaller number of groups such that data points in the same group share the same properties while, in different groups, they are dissimilar. Clustering has many applications, including part family formation for group technology, image segmentation, information retrieval, web pages grouping, market segmentation, and scientific and engineering analysis (16). (
CUSTOMER DATA CLUSTERING USING DATA MINING TECHNIQUE)
SOA reliability is very important, but there are problems that can be cause system failure. Fixing errors in the system like, detecting bugs ca take time and can be difficult. The large quantity of analysis data in large software such as source code and documents, however, renders a tedious and difficult task on developers to analyze them.
"One of the findings in bug characteristic study is that semantic error is the major root cause of bugs in modern software. Semantic bugs are application specific and so it requires knowledge about the application to detect them. To address this problem, this dissertation proposes using data mining technique to automatically detect software bugs. To demonstrate this approach, this dissertation presents two automatic bug detection tools, including PR-Miner that extracts programming rules and detects violations, and CP-Miner that detects copy-pasted code and related bugs. Using Data Mining Techniques to Improve Software Reliability"
Large software use data mining techniques, such as PR-miner, as a bug detection tool. Programs follow, that are documented manually by programmers. Bugs can be introduced if rules are violated. "PR-Miner uses itemset mining to efficiently extract programming rules from large software code, requiring little effort from programmers and no prior knowledge of the softwareUsing Data Mining Techniques to Improve Software Reliability". Programmers can benefit from PR-Miner by extracting programm rules that contain different elements of multiple types. An efficient algorithm to automatically detect violations to the extracted programming rules, which are strong indications of bugs. PR-Miner can efficiently extract thousands of general programming rules and detect violations within minutes. Moreover, PR-Miner has detected many violations to the extracted rules, which are potential bugs.
One consumer issue can be unathorized usage of consumers information. Agency may collect information that is unauthorized and the consumers privacy is invaded. Information held by third parties can be accessed by the government. Questions are raised why the government would need this information. Would it be used as a government source, commercial source or both. Issues that may come up is the use of