Essay Example on The objective of mobile data mining is to provide efficient Techniques

The objective of mobile data mining is to provide efficient techniques for the analysis and monitoring of critical data from mobile devices The client programs are present on mobile devices that can invoke the data mining tasks of remote execution and show the mining results is an additional value for organizations and users who need to perform data analysis stored where users are working which is very far away from the site that allows them to produce study regardless of their present location The mobile data mining provides efficient techniques for the monitoring of critical data from mobile devices Distributed data mining environment is a critical issue which mobile data mining has to face also some technological constraints such as low bandwidth of networks much slower processors lesser storage space small screens to visualize the results as well as limited battery power MobiMine is an example of mobile data mining environment that is created for intelligent supervising from mobile devices of stock market It is based on a client server architecture 

The clients which are running such as PDAs monitor a stream of all the data which are coming through a server The stock market data is collected from number of Web sources in a database and evaluate it on a regular basis using several mining techniques The database is queried by clients for the latest information as for quotes and other essential information For communication which includes clients and the database a proxy is used When a user have some query regarding the database to the proxy the query is send that connects to the database take out the results and pass them to the client MobiMine uses a Fourier dependent approach for the representation of the decision trees for efficient communication of data mining models over wireless links that has limited Bandwidth it saves memory on mobile device and bandwidth of network These are the major components of any data mining system data sources data warehouse server engine for data mining graphical user interface GUI pattern evaluation module and knowledge base System architecture of mobile data mining is based on three types of basic components 1 Data Sources Data warehouse World Wide Web Database some text files and other documents which are the actual sources of data present We require large volumes of historic data to be successful for data mining Organizations mostly store data in data warehouses It may contain one or more databases files spreadsheets or other kinds of information database Mostly data may present even in plain text files and spreadsheets The another big source of data is World Wide Web or the Internet Different Processes 

Whenever the getStatus operation return done the MIDlet invokes the getResult operation to receive the result of the data mining analysis 3 Mining Server Server nodes are essential for storing the data generated by some data providers and for executing the mining tasks submitted by mobile clients Each mining server uncovers its functionalities by the help of two web services Data Collection Service DCS 1 The DCS is invoked by data providers to store data on the server 2 For uploading new data 3 For deleting an existing data 4 For updating data set Data Mining Service DMS 1 Invoked by mobile clients to perform data mining tasks Obtaining the list of the available data sets and algorithms 2 Submitting a data mining task 3 Getting the current status of a computation along with the result of a given task 4 The DCS is invoked by data providers to store data on the server Data uploaded through the DCS is stored as plain data sets in the local file system The DMS invokes the mobile clients to perform data mining tasks Its interface defines a set of operations DMS ops The data analysis is performed by the DMS using a subset of all the algorithms