Aiming at some existing issues in the sports industry, the existing model is optimized by deep learning and time series theory based on the relevant algorithm, and the scale of the sports industry is analyzed and predicted by the model. The results show the following: (1) Based on the single-step prediction of time series, MNTS structural algorithm can be used to describe and study the sports industry scale with the single factor, and the correlation fitting degree is high. (2) Curves of different evaluation methods can include parts linear stage and nonlinear stage according to the magnitude of change. (3) Seen from the optimization model in this paper, the proposed method can describe both global and local trends of data. (4) It can be seen from the prediction curve that the overall state of fluctuation indicates that time will have a great impact on the relevant scale of the sports industry. Compared with single-step prediction, the accuracy of multistep prediction is higher, and the multistep prediction model based on time series can well characterize and predict the scale of the sports industry. By using the relevant time algorithm, the sports industry scale can be predicted and analyzed so as to provide theoretical support for the formulation and implementation of relevant policies.

/pdf/an-intelligent-prediction-for-sports-industry-scale-based-on-1nvftmaf.pdf

An Intelligent Prediction for Sports Industry Scale Based on Time Series Algorithm and Deep Learning

Dictionary-based classifiers are an essential group of approaches in the field of time series classification. Their distinctive characteristic is that they transform time series into segments made of symbols (words) and then classify time series using these words. Dictionary-based approaches are suitable for datasets containing time series of unequal length. The prevalence of dictionary-based methods inspired the research in this paper. We propose a new dictionary-based classifier called SAFE. The new approach transforms the raw numeric data into a symbolic representation using the Simple Symbolic Aggregate approXimation (SAX) method. We then partition the symbolic time series into a sequence of words. Then we employ the word embedding neural model known in Natural Language Processing to train the classifying mechanism. The proposed scheme was applied to classify 30 benchmark datasets and compared with a range of state-of-the-art time series classifiers. The name SAFE comes from our observation that this method is safe to use. Empirical experiments have shown that SAFE gives excellent results: it is always in the top 5%–10% when we rank the classification accuracy of state-of-the-art algorithms for various datasets. Our method ranks third in the list of state-of-the-art dictionary-based approaches (after the WEASEL and BOSS methods). 

Time-series classification with SAFE: Simple and fast segmented word embedding-based neural time series classifier

https://link.springer.com/content/pdf/10.1007/s10115-021-01617-w.pdf

ProAID: path-based reasoning for self-attentional disease prediction

With the rapid development of sensor components, a large number of time series have been generated by various sensors, and many time series classification (TSC) algorithms have been proposed to process these data. Among them, the algorithm based on hapelet attracts much attention because of its high accuracy and explanability. However, the high time complexity of the algorithm based on Shapelet and the poor quality of the selected Shapelet limit the accuracy and running time of the classification. This paper presents an improved time series classification algorithm based on Shapelet transformation (BSPST), which greatly reduces the time for Shapelet selection and greatly improves the quality of Shapelet discrimination. A high-quality set of candidate Shapelets is preserved to build classification models using Bloom filters and similarity matching. Data is converted using Shapelet conversion technology. Sufficient experimental validation was performed with 44 common datasets. Testing the classification accuracy and running time, experiments show that BSPST keeps a high accuracy and is significantly faster than the existing Shapelet selection algorithm. It effectively solves the problem of time series classification for a variety of sensor instruments.

BSPST:A classification algorithm based on shape transformation

With the rapid development of sensor components, a large number of time series have been generated by various sensors, and many time series classification (TSC) algorithms have been proposed to process these data. Among them, the algorithm based on hapelet attracts much attention because of its high accuracy and explanability. However, the high time complexity of the algorithm based on Shapelet and the poor quality of the selected Shapelet limit the accuracy and running time of the classification. This paper presents an improved time series classification algorithm based on Shapelet transformation (BSPST), which greatly reduces the time for Shapelet selection and greatly improves the quality of Shapelet discrimination. A high-quality set of candidate Shapelets is preserved to build classification models using Bloom filters and similarity matching. Data is converted using Shapelet conversion technology. Sufficient experimental validation was performed with 44 common datasets. Testing the classification accuracy and running time, experiments show that BSPST keeps a high accuracy and is significantly faster than the existing Shapelet selection algorithm. It effectively solves the problem of time series classification for a variety of sensor instruments. 

In the research area of time series classification, the ensemble shapelet transform algorithm is one of the state-of-the-art algorithms for classification. However, its high time complexity is an issue to hinder its application since its base classifier shapelet transform includes a high time complexity of a distance calculation and shapelet selection. Therefore, in this paper we introduce a novel algorithm, i.e., short isometric shapelet transform (SIST), which contains two strategies to reduce the time complexity. The first strategy of SIST fixes the length of shapelet based on a simplified distance calculation, which largely reduces the number of shapelet candidates as well as speeds up the distance calculation in the ensemble shapelet transform algorithm. The second strategy is to train a single linear classifier in the feature space instead of an ensemble classifier. The theoretical evidence of these two strategies is presented to guarantee a near-lossless accuracy under some preconditions while reducing the time complexity. Furthermore, empirical experiments demonstrate the superior performance of the proposed algorithm.

Short isometric shapelet transform for binary time series classification

In the research area of time series classification, the ensemble shapelet transform algorithm is one of state-of-the-art algorithms for classification. However, its high time complexity is an issue to hinder its application since its base classifier shapelet transform includes a high time complexity of a distance calculation and shapelet selection. Therefore, in this paper we introduce a novel algorithm, i.e. short isometric shapelet transform, which contains two strategies to reduce the time complexity. The first strategy of SIST fixes the length of shapelet based on a simplified distance calculation, which largely reduces the number of shapelet candidates as well as speeds up the distance calculation in the ensemble shapelet transform algorithm. The second strategy is to train a single linear classifier in the feature space instead of an ensemble classifier. The theoretical evidences of these two strategies are presented to guarantee a near-lossless accuracy under some preconditions while reducing the time complexity. Furthermore, empirical experiments demonstrate the superior performance of the proposed algorithm.

Weibo Shu

Papers

Short isometric shapelet transform for binary time series classification

Use Short Isometric Shapelets to Accelerate Binary Time Series Classification