Geodesic
A method for representing stock time series features based on trend and inclination angle turning points
Computer Science and Information Systems, Tome 23 (2026) no. 1

Voir la notice de l'article provenant de la source Computer Science and Information Systems website

Stock time series data mining faces significant challenges in terms of time and space. Time series feature representation is an important means of reducing the dimensionality of time series. This study proposes a stock time series feature representation method based on a combination of slope angle changes and trend turning point screening. This method alleviates the limitations of single methods, such as local feature overfitting, loss of global trends, and sparse long-segment features, through a multi-level collaborative mechanism of screening, filtering, and supplementation. Experimental results show that compared with four other methods, the proposed method effectively preserves the original features of stock time series, achieving good results in both similarity metrics and fitting errors.
Keywords: stock time series, feature representation, inclination angle turning point, trend analysis 
Lei Han; Xuedong Gao. A method for representing stock time series features based on trend and inclination angle turning points. Computer Science and Information Systems, Tome 23 (2026) no. 1. http://geodesic.mathdoc.fr/item/CSIS_2026_23_1_a9/
@article{CSIS_2026_23_1_a9,
     author = {Lei Han and Xuedong Gao},
     title = {A method for representing stock time series features based on trend and inclination angle turning points},
     journal = {Computer Science and Information Systems},
     year = {2026},
     volume = {23},
     number = {1},
     url = {http://geodesic.mathdoc.fr/item/CSIS_2026_23_1_a9/}
}
TY  - JOUR
AU  - Lei Han
AU  - Xuedong Gao
TI  - A method for representing stock time series features based on trend and inclination angle turning points
JO  - Computer Science and Information Systems
PY  - 2026
VL  - 23
IS  - 1
UR  - http://geodesic.mathdoc.fr/item/CSIS_2026_23_1_a9/
ID  - CSIS_2026_23_1_a9
ER  - 
%0 Journal Article
%A Lei Han
%A Xuedong Gao
%T A method for representing stock time series features based on trend and inclination angle turning points
%J Computer Science and Information Systems
%D 2026
%V 23
%N 1
%U http://geodesic.mathdoc.fr/item/CSIS_2026_23_1_a9/
%F CSIS_2026_23_1_a9