Publication

Abstract : Pattern recognition is crucial across diverse domains, including retrieval of information, data mining, and bioinformatics. Numerous algorithms exist for string matching, and finite state machines (FSM) provide a robust framework for implementing these pattern recognition algorithms. This study introduces an efficient pattern recognition application utilizing a finite state machine, using the SSFSM library in Python. The SSFSM library facilitates the creation and management of FSMs, offering methods to define states and transitions efficiently. An algorithm is proposed, that identifies strings within a text file with the help of the SSFSM library in Python. It employs deterministic finite automata (DFA) for pattern recognition, scanning text files word by word and then character by character, and recording execution times to evaluate efficiency. The algorithm’s effectiveness is demonstrated through a comparative analysis of its performance for the number of characters and size of the input file. Results demonstrate the algorithm’s effectiveness in accurately identifying and quantifying specific string occurrences within large text files. It also shows that the search time of a string increases with an increase in the frequency of the character, size of the text file, and length of the search key, underscoring its potential for broader applications. Further, our results also show that there is approximately a 20% increase in search time when the search key length increases by 1. Future enhancements may include extended functionalities, support for additional document types, and improved complexity management.