Research And Implementation Of Natural Language Processing Algorithm Based On Smart Home

Smart Home is a research hotspot which receives wide attention in recent years. With the wave of Internet +, the research process on Smart Home was pushed to a new height, the requirement of intelligence also become higher. Smart home research difficulties are not only focused on the technical level, but on comfortable, humanistic ideas. From the traditional button control development up to now, people began to expect to realize the control of household equipment through language. In this case, the combination of NLP and Smart Home received more attention.Natural language processing is mainly divided into two parts: the speech processing, and word processing. This article does not consider the phonetic level, but from the aspect of word processing is analyzed. At the level of word processing, Natural Language Processing is divided into three parts: lexical analysis, syntactic analysis and semantic analysis. Involved in the field of intelligent home language ambiguity and lexical scope, use the maximum matching algorithm for segmentation of natural language. Then, according to the characteristics of the field of intelligent home, choose the phrase structure method is used to analyze the grammatical structure. Finally, the template matching method for semantic understanding, and ultimately the formation of the code to control the node. In this paper the application of the system, over the phone has been the voice of the input feature will be processed into text, after the language processing method to get the control instruction. The processed data through the socket communication with the server, the server will send the data to the node through the Zigbee.This paper has completed the following tasks in this respect:(1) On the above areas of the relevant methods and knowledge to understand, the natural language is divided into word segmentation and tagging part of speech, grammatical level analysis, semantic understanding of the three parts of the level of analysis. Each part of the relevant methods to collect, combined with the specific circumstances of the field of intelligent home, from which to select the most suitable analysis method, and to modify the word segmentation algorithm;(2)According to the understanding of the field of intelligent home, combined with specific situation, has established the field of intelligent home special lexicon, syntax database, semantic template matching, to support natural language understanding the realization of the algorithm;( 3) Implementation and integration of lexical analysis, syntactic analysis and semantic analysis algorithm, formed a complete semantic understanding method, and design between semantics and response generation method, and into a digital form that the computer can understand natural language processing, the final completion of the smart home control;(4)Finally, the improved segmentation algorithm is tested, and the accuracy of the algorithm is tested and applied to the smart home mobile client. The feasibility of the algorithm is verified.