Extracting The Best Features For Predicting The Trend Of Tehran Stock Exchange (TSE)
کد مقاله : 1100-CFIS (R1)
نویسندگان:
فرزانه اکبرزاده *1، علی سلیمانی ایوری2
1دانشکده مهندسی کامپیوتر، دانشگاه صنعتی شاهرود، شاهرود، ایران
2دانشکده مهندسی کامپیوتر، دانشگاه صنعتی شاهرود، شاهرود، ایران
چکیده مقاله:
Predicting stock price trend is an important task as well as difficult problem. This prediction depends on various factors and their complex relationships. Feature extraction is a way to discover these factors from initial data. In this paper, several famous feature extraction methods are proposed as a data engineering approach for TSE. The results of predicting next day trend by k-Nearest Neighbors (kNN) predictor, shows indicators are best extracted features.
کلیدواژه ها:
stock, trend prediction, feature extraction method, kNN
وضعیت : مقاله برای ارائه شفاهی پذیرفته شده است