A Framework of Data Mining for Wireless Sensor Network based Applications

Abstract

newline One of the most exciting new developments in computer science is the study of wireless sensor newlinenetworks (WSN). A wireless sensor network (WSN) is a group of sensors used to monitor the newlineenvironment that aren t necessarily located in one particular place. These featherweight sensors newlinewere made to keep an eye on and manage cutting-edge machinery. Making the ongoing decision newlinehas emerged as the most important portion of constructing the WSNs-built applications due to newlineabsolute asset constrained processing, imparting constraints, and the enormous volume of newlinecontinuously changing information offered by WSNs. To efficiently filter through this mountain of newlinedata and discover important trends and patterns, it is becoming clear that a cutting-edge yet widely newlinerecognized data mining approach is required. If people had more time to prepare for natural newlinedisasters like landslides, earthquakes, floods, forest fires, tsunamis, etc., more lives may be saved. newlineLife can be saved by keeping an eye on disaster areas and alerting the public. Data Mining, the newlinepractice of extracting relevant information from a large, well organized database, can be used in newlinemany contexts. In order to maximize the usefulness of the sensor node, we plan to employ data newlinemining techniques. The primary objective of this study was to lay the groundwork for future newlinestudies analyzing the current status of data mining in WSN. newlineIn this study, forecasts are made using the proposed models. Intel s research lab used 54 sensors to newlineset up a wireless sensor network, cutting-edge networking technology at the time, to gather the newlineinformation needed to develop the solution. As inputs, the model takes into account a wide range newlineof meteorological factors, allowing for more accurate weather forecasting and disaster monitoring. newlineIn addition to predict water quality, considered water dataset with the key water parameters as newlinetemperature, DO, PH, conductivity, BOD, Nitratenan, FecalColiform, and TotalColiform. newlineThe model is assessed by means of the Mean Absolute Error (MAE), Root M