The Most Advantageous Bangla Keyboard Layout Using Data Mining Technique

Bangla alphabet has a large number of letters, for this it is complicated to type faster using Bangla keyboard. The proposed keyboard will maximize the speed of operator as they can type with both hands parallel. Association rule of data mining to distribute the Bangla characters in the keyboard is used here. The frequencies of data consisting of monograph, digraph and trigraph are analyzed, which are derived from data ire-house, and then used association rule of data mining to distribute the Bangla characters in the layout. ow the effectiveness of the proposed approach with better l Bangla Keyboard Layout, which distributes the load equally on both hands so that maximizing the ease and minimizing the effort. Monograph, digraph, trigrap itching, support, confidence. mputer is increasing rapidly. It is the fastest growing industry in the world. New invention of technologies makes it faster. The billions of instructions can be done within a second now a days. But the technology of input device like keyboard has not been changed much. The typing speed is very slow in respect of computer’s other devices. Its main mplex and insufficient keyb an limitations [8-9]. A scientific t with equal hands load and maximum hand switching can reduce this problem. The use of Bangla is increasing day by day in everyday life. Specially in data entry and printing sector. But there is no scientific Bangla keyboard layout at present and very few research works have been done in this field. We use the concept of data mining association rule to design Bangla keyboard layout that shows optimal and better performance than the existing keyboard layout. design a Bangla keyboard layout using the association rule of data mining. We have collected the data from various Bangla documents and we have used the association rule to extract the association of Bangla letters each other. And finally we have designed a Bangla keyboard layout with equal hands load and maximum hand switching.

[1]  Heikki Mannila,et al.  Fast Discovery of Association Rules , 1996, Advances in Knowledge Discovery and Data Mining.

[2]  Heikki Mannila,et al.  Verkamo: Fast Discovery of Association Rules , 1996, KDD 1996.

[3]  Jiawei Han,et al.  A fast distributed algorithm for mining association rules , 1996, Fourth International Conference on Parallel and Distributed Information Systems.

[4]  Jian Pei,et al.  Mining frequent patterns without candidate generation , 2000, SIGMOD '00.

[5]  Jian Pei,et al.  Mining frequent patterns without candidate generation , 2000, SIGMOD 2000.