الفهرس 
Table of contentsOnly 14 pages are availabe for public view
 |  | from | 157 |  |  |
| | | from | 157 | | |
Abstract
Information technology contributes to the development of various fields. Data mining or knowledge discovery applies various algorithms to discover information and extract patterns of stored data. Data mining has significant applications for decision making. Its techniques affect Databases, Statistics, Pattern Reorganization, Machine Learning, Artificial Intelligence and Computation capabilities. Therefore, data mining assumes a vital part in the search for information and knowledge discovery. It incorporates an alternate assortment of data with a specific end goal to accomplish. So, data mining techniques used the term Text Mining for the Discovery and analysis of text. The difference between text mining and data mining depends on the source of data. In text mining, fundamentally, the data input is the unstructured file while for data mining the input is of structured data. The idea is that patterns are extracted from unstructured text in text mining, while in data mining structured data is used. Speech Recognition (SR) is the interpretation of talked words into text. It is, otherwise, called ”Automatic Speech Recognition (ASR) ”, ”computer speech recognition”, ”Speech to Text”, or ”STT”. Speech is a characteristic method of communication for people and it passes on some information. This information can be utilized for various purposes like confirmation, text conversion or machine control based on the application. People feel so good to connect with computers by speech, instead of falling back on primitive interfaces, for example, keyboards and pointing devices. Keyword spotting (KWS) is a mechanically significant issue, assuming an essential part in sound indexing and speech data mining applications. KWS is also utilized for finding events of the keyword in speech signal. This issue is like speech recognition, but the additional signals around the words of interest must be disregarded. As an essential part in the automatic Speech Recognition (ASR) components, KWS system converts the speech into text. The converted text is used as the information source of indexing part for the KWS, so the reliability of ASR components highly affects the performance of KWS systems. If the recognition result is mostly correct, the KWS system works well in comparison with the result of ASR is used.