Cancer Prediction by Data Mining

By definition, data mining means finding patterns in large data sets which can then be used for analysis of data. This analysis can be used for getting more customers or clients, generating an increase in revenues compared to that of last year. Now how does this actually work? For easy and effective solutions, read on.

Data Mining is currently used in fraud detection of credit card usage. A person who uses a credit card in different places is saved and the pattern is generated using a data mining tool. When there is an exception or outlier for the credit card usage in a far off distant place, the tool will help to detect the pattern, catch the fraudulent person, and the credit card’s user is informed and his money restored.

Approximate string matching algorithms can be used as data mining tools and applied to a data set of strings to retrieve an approximate closely matched string in response to a query string posed to the system. In a similar way, a query finger print can be posed to a data set of saved finger prints of criminals and if there is a match, the criminal is identified and penalized for any recent crimes.

Text mining can be carried out in online bookshop systems. A text mining tool is executed on the reviews of a book and the comments in the reviews are analyzed for sentiment values – bad, moderate, good or excellent using the tool and the results can be analyzed to find out patterns such as which books are in great demand and should be brought in more supplies to meet customers’ requirements and hence purchases, increasing the profits from the system.

Coming to talk of cancer prediction by data mining, it works in a somewhat similar manner. The patient in emergency cases when the right doctor is unavailable for instance, during holiday time or in the dead of the night is facing some symptoms and is eager to know whether it has anything to do with cancer. He can just open an end user application which is like an online questionnaire or survey, posing questions related to the symptoms of the cancer disease or any other related diseases along with cancer and will help the symptoms-suffering person to understand if he at all has cancer or, if he does have cancer, what other related diseases he has.

The online application after completion and hitting submit will directly take him to the result report page and if he is found to be cancer positive, the patient will also be able to view a list of available doctors he should contact along with their detailed profiles so that he can choose the right doctor and make an appointment and contact in person for real-time tests and further diagnosis.

Summing up, in these ways, data mining tools, by finding patterns in data sets can help in a wide range of areas and they get more popular as more research is carried out and implemented in real-time for customers, clients and patients to give them high-ranking service.