Data mining & learning techniques 4 future network knowledge

my project name is data mining and learning techniques for future network knowledge management i want some one who is expert in writing thesis for me. i have done my practical work upto 70% but need some one to write thesis for me my project details are as follows Data Mining and Learning techniques for future Network knowledge Management Abstract Autonomic network management in the future networks helps the network operators in reducing the Operational Costs (OPEX) incurred by manually management of the networks by human operators. One of the basic requirements of autonomic networking is awareness of the status of the network, observing the faults and resources and automatically configuring the network based on the information gathered from the system. The auto configuration mentioned above, cannot always be performed based on of the raw information taken from the network, and some techniques such as data mining and learning algorithms are required to prepare the data into a prepared piece of “Knowledge” in the future networks as data mining is a new and powerful technology with the ability and potential used for extracting of analytical and predictive information from huge databases and is widely helpful for companies to focus on the important information in their data warehouses. Proposed scenario At the heart of the telecom industry is the BTS. BTS outages are a routine happening and backup plans are almost always in place. However, outages lead to downgrading of performance and costly repair. It would be helpful if an automated, intelligent system was in place to provide early-failure warnings and the expected point of failure. To achieve this, there is a need to data-mine the outage data and co-relate the outages with different possible causes, including weather and atmospheric conditions, traffic load, number of connected users, time of day and servicing history. The data would also need to be co-related with the diagnostic report for outage so that the expected point of failure can also be identified along with the warning that is generated. To achieve this data needs to be collected from metrological department for weather and atmosphere. BTS outage and diagnostic information should also be available. Data regarding servicing and maintenance of BTS will also be required. Also, if possible the network snapshot before and at the time of outage should be available. This data shall be cleaned via Google Refine, which provides powerful options for normalizing data. Then the data shall be loaded via SSIS into SQL Server and cubes shall be generated. Once the data is available in data warehouse in form of cubes, appropriate queries and slicing and dicing can be performed to figure out what are most probable scenarios for BTS outage. This ‘knowledge’ can be used to warn of possible outages and probably prevent failures before they occur.