| 1. |
- Sun, Bin, et al.
(author)
-
An Improved k-Nearest Neighbours Method for Traffic Time Series Imputation
- 2017
-
Conference paper (peer-reviewed)abstract
- Intelligent transportation systems (ITS) are becoming more and more effective, benefiting from big data. Despite this, missing data is a problem that prevents many prediction algorithms in ITS from working effectively. Much work has been done to impute those missing data. Among different imputation methods, k-nearest neighbours (kNN) has shown excellent accuracy and efficiency. However, the general kNN is designed for matrix instead of time series so it lacks the usage of time series characteristics such as windows and weights that are gap-sensitive. This work introduces gap-sensitive windowed kNN (GSW-kNN) imputation for time series. The results show that GSW-kNN is 34% more accurate than benchmarking methods, and it is still robust even if the missing ratio increases to 90%.
|
|
| 2. |
- Jin, Binghan, et al.
(author)
-
Abdominal Adiposity and Total Body Fat as Predictors of Cardiometabolic Health in Children and Adolescents With Obesity
- 2020
-
In: Frontiers in Endocrinology. - : FRONTIERS MEDIA SA. - 1664-2392. ; 11
-
Journal article (peer-reviewed)abstract
- Objective:We aimed to assess the role of adipose tissue distribution in cardiometabolic risk (in particular insulin sensitivity) in a population of children and adolescents with obesity. Methods:In this cross-sectional study, participants were 479 children and adolescents with obesity (322 boys and 157 girls) aged 3 to 18 years attending the Children's Hospital at Zhejiang University School of Medicine (Hangzhou, China). Clinical assessments included anthropometry, body composition (DXA scans), carotid artery ultrasounds, and OGTT. Insulin sensitivity was assessed using the Matsuda index. Participants were stratified into groups by sex and pubertal stage. Key predictors were DXA-derived android-to-gynoid-fat ratio (A/G) and total body fat percentage (TBF%). Results:Irrespective of sex and pubertal stage, there was a strong association between increasing A/G (i.e., greater abdominal adiposity) and lower insulin sensitivity. In multivariable models, every 0.1 increase in A/G was associated with a reduction in insulin sensitivity in prepubertal boys [-29% (95% CI -36%, -20%);p< 0.0001], pubertal boys [-13% (95% CI -21%, -6%);p= 0.001], and pubertal girls [-16% (95% CI -24%, -6%);p= 0.002]. In contrast, TBF% was not associated with insulin sensitivity when A/G was adjusted for, irrespective of pubertal stage or sex. In addition, every 0.1 increase in A/G was associated with increased likelihood of dyslipidemia in prepubertal boys [adjusted odds ratio (aOR) 1.62 (95% CI 1.05, 2.49)], impaired glucose tolerance in pubertal boys [aOR 1.64 (95% CI 1.07, 2.51)] and pubertal girls [aOR 1.81 (95% CI 1.10, 2.98)], and odds of NAFLD in both prepubertal [aOR 2.57 (95% CI 1.56, 4.21)] and pubertal [aOR 1.69 (95% CI 1.18, 2.40)] boys. In contrast, higher TBF% was only associated with higher fasting insulin and ALT in pubertal boys, being also predictive of NAFLD in this group [aOR 1.15 per percentage point (95% CI 1.06, 1.26)], but was not associated with the likelihood of other cardiometabolic outcomes assessed in any group. Conclusions:A/G is a much stronger independent predictor of cardiometabolic risk factors in children and adolescents with obesity in China, particularly glucose metabolism.
|
|
| 3. |
- Liu, Jian, et al.
(author)
-
Design of Full-Scale Endwall Film Cooling of a Turbine Vane
- 2020
-
In: Journal of Heat Transfer. - : ASME International. - 0022-1481 .- 1528-8943. ; 142:2
-
Journal article (peer-reviewed)abstract
- Endwall film cooling is a significant cooling method to protect the endwall region and the junction region of endwall and a turbine vane, where usually a relatively high temperature load exists. This work aims to find the optimized arrangement of film cooling holes on the endwall and improve the film cooling in some difficult regions on the endwall, such as pressure side-endwall junction region. Several ideas for film cooling hole arrangement design are proposed, based on the pressure coefficient distribution, the streamline distribution, and the heat transfer coefficient (HTC) distribution, respectively. Four specified designs are built and compared. The results are obtained by numerical calculations with a well-validated turbulence model, the k-ω shear stress transport (SST) model. From this work, the designs based on the pressure coefficient distribution (designs 1 and 2) force the flow from the pressure side to the suction side (SS), especially in design 2, which adopts compound angle holes. The designs based on pressure coefficients have benefit in the cooling of the SS but give worse coolant coverage on the pressure side. In addition, designs 1 and 2 have little influence on the original pressure field. The design based on the streamline distributions (design 3) has larger coolant coverage on the endwall and provides good coolant coverage on the endwall and pressure side junction region. The design based on the HTC distribution provides large overall film cooling effectiveness on both the pressure side and the SS. More film cooling holes are placed on the high temperature regions, which is more effective in practice.
|
|
| 4. |
- Qu, Guangbo, et al.
(author)
-
Identification of tetrabromobisphenol A allyl ether and tetrabromobisphenol A 2,3-dibromopropyl ether in the ambient environment near a manufacturing site and in mollusks at a coastal region
- 2013
-
In: Environmental Science and Technology. - : American Chemical Society (ACS). - 0013-936X .- 1520-5851. ; 47:9, s. 4760-4767
-
Journal article (peer-reviewed)abstract
- Tetrabromobisphenol A (TBBPA) is one of the most widely used brominated flame retardants (BFRs) and has been frequently detected in the environment and biota. Recent studies have found that derivatives of TBBPA, such as TBBPA bis(allyl) ether (TBBPA BAE) and TBBPA bis(2,3-dibromopropyl) ether (TBBPA BDBPE) are present in various environmental compartments. In this work, using liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) and liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (LC-Q-TOF-MS), TBBPA allyl ether (TBBPA AE) and TBBPA 2,3-dibromopropyl ether (TBBPA DBPE) were identified in environmental samples and further confirmed by synthesized standards. Soil, sediment, rice hull, and earthworm samples collected near a BFR manufacturing plant were found to contain these two compounds. In sediments, the concentrations of TBBPA AE and TBBPA DBPE ranged from 1.0 to 346.6 ng/g of dry weight (dw) and from 0.7 to 292.7 ng/g of dw, respectively. TBBPA AE and TBBPA DBPE in earthworm and rice hull samples were similar to soil samples, which ranged from below the method limit of detection (LOD, <0.002 ng/g of dw) to 0.064 ng/g of dw and from below the LOD (<0.008 ng/g of dw) to 0.58 ng/g of dw, respectively. Furthermore, mollusks collected from the Chinese Bohai Sea were used as a bioindicator to investigate the occurrence and distribution of these compounds in the coastal environment. The detection frequencies of TBBPA AE and TBBPA DBPE were 41 and 32%, respectively, and the concentrations ranged from below LOD (<0.003 ng/g of dw) to 0.54 ng/g of dw, with an average of 0.09 ng/g of dw, for TBBPA AE, and from below LOD (<0.008 ng/g of dw) to 1.41 ng/g of dw, with an average of 0.15 ng/g of dw, for TBBPA DBPE.
|
|
| 5. |
- Sun, Bin, et al.
(author)
-
An Overview of Parameter and Data Strategies for K-Nearest Neighbours Based Short-Term Traffic Prediction
- 2017
-
In: ACM International Conference Proceeding Series Volume Part F133326. - New York, NY, USA : Association for Computing Machinery (ACM). - 9781450353762 ; , s. 68-74
-
Conference paper (peer-reviewed)abstract
- Modern intelligent transportation systems (ITS) requires reliable and accurate short-term traffic prediction. One widely used method to predict traffic is k-nearest neighbours (kNN). Though many studies have tried to improve kNN with parameter strategies and data strategies, there is no comprehensive analysis of those strategies. This paper aims to analyse kNN strategies and guide future work to select the right strategy to improve prediction accuracy. Firstly, we examine the relations among three kNN parameters, which are: number of nearest neighbours (k), search step length (d) and window size (v). We also analysed predict step ahead (m) which is not a parameter but a user requirement and configuration. The analyses indicate that the relations among parameters are compound especially when traffic flow states are considered. The results show that strategy of using v leads to outstanding accuracy improvement. Later, we compare different data strategies such as flow-aware and time-aware ones together with ensemble strategies. The experiments show that the flowaware strategy performs better than the time-aware one. Thus, we suggest considering all parameter strategies simultaneously as ensemble strategies especially by including v in flow-aware strategies.
|
|
| 6. |
- Sun, Bin, 1988-
(author)
-
Automated Traffic Time Series Prediction
- 2018
-
Doctoral thesis (other academic/artistic)abstract
- Intelligent transportation systems (ITS) are becoming more and more effective. Robust and accurate short-term traffic prediction plays a key role in modern ITS and demands continuous improvement. Benefiting from better data collection and storage strategies, a huge amount of traffic data is archived which can be used for this purpose especially by using machine learning.For the data preprocessing stage, despite the amount of data available, missing data records and their messy labels are two problems that prevent many prediction algorithms in ITS from working effectively and smoothly. For the prediction stage, though there are many prediction algorithms, higher accuracy and more automated procedures are needed.Considering both preprocessing and prediction studies, one widely used algorithm is k-nearest neighbours (kNN) which has shown high accuracy and efficiency. However, the general kNN is designed for matrix instead of time series which lacks the use of time series characteristics. Choosing the right parameter values for kNN is problematic due to dynamic traffic characteristics. This thesis analyses kNN based algorithms and improves the prediction accuracy with better parameter handling using time series characteristics.Specifically, for the data preprocessing stage, this work introduces gap-sensitive windowed kNN (GSW-kNN) imputation. Besides, a Mahalanobis distance-based algorithm is improved to support correcting and complementing label information. Later, several automated and dynamic procedures are proposed and different strategies for making use of data and parameters are also compared.Two real-world datasets are used to conduct experiments in different papers. The results show that GSW-kNN imputation is 34% on average more accurate than benchmarking methods, and it is still robust even if the missing ratio increases to 90%. The Mahalanobis distance-based models efficiently correct and complement label information which is then used to fairly compare performance of algorithms. The proposed dynamic procedure (DP) performs better than manually adjusted kNN and other benchmarking methods in terms of accuracy on average. What is better, weighted parameter tuples (WPT) gives more accurate results than any human tuned parameters which cannot be achieved manually in practice. The experiments indicate that the relations among parameters are compound and the flow-aware strategy performs better than the time-aware one. Thus, it is suggested to consider all parameter strategies simultaneously as ensemble strategies especially by including window in flow-aware strategies.In summary, this thesis improves the accuracy and automation level of short-term traffic prediction with proposed high-speed algorithms.
|
|
| 7. |
- Sun, Bin, 1988-, et al.
(author)
-
Correcting and complementing freeway traffic accident data using mahalanobis distance based outlier detection
- 2017
-
In: Technical Gazette. - : Strojarski Facultet. - 1330-3651 .- 1848-6339. ; 24:5, s. 1597-1607
-
Journal article (peer-reviewed)abstract
- A huge amount of traffic data is archived which can be used in data mining especially supervised learning. However, it is not being fully used due to lack of accurate accident information (labels). In this study, we improve a Mahalanobis distance based algorithm to be able to handle differential data to estimate flow fluctuations and detect accidents and use it to support correcting and complementing accident information. The outlier detection algorithm provides accurate suggestions for accident occurring time, duration and direction. We also develop a system with interactive user interface to realize this procedure. There are three contributions for data handling. Firstly, we propose to use multi-metric traffic data instead of single metric for traffic outlier detection. Secondly, we present a practical method to organise traffic data and to evaluate the organisation for Mahalanobis distance. Thirdly, we describe a general method to modify Mahalanobis distance algorithms to be updatable. © 2017, Strojarski Facultet. All rights reserved.
|
|
| 8. |
- Sun, Bin, et al.
(author)
-
Flow-Aware WPT k-Nearest Neighbours Regression for Short-Term Traffic Prediction
- 2017
-
In: Proceedings - IEEE Symposium on Computers and Communications. - : Institute of Electrical and Electronics Engineers (IEEE). - 9781538616291 ; , s. 48-53
-
Conference paper (peer-reviewed)abstract
- Robust and accurate traffic prediction is critical in modern intelligent transportation systems (ITS). One widely used method for short-term traffic prediction is k-nearest neighbours (kNN). However, choosing the right parameter values for kNN is problematic. Although many studies have investigated this problem, they did not consider all parameters of kNN at the same time. This paper aims to improve kNN prediction accuracy by tuning all parameters simultaneously concerning dynamic traffic characteristics. We propose weighted parameter tuples (WPT) to calculate weighted average dynamically according to flow rate. Comprehensive experiments are conducted on one-year real-world data. The results show that flow-aware WPT kNN performs better than manually tuned kNN as well as benchmark methods such as extreme gradient boosting (XGB) and seasonal autoregressive integrated moving average (SARIMA). Thus, it is recommended to use dynamic parameters regarding traffic flow and to consider all parameters at the same time.
|
|
| 9. |
- Sun, Bin, 1988-, et al.
(author)
-
Short-Term Traffic Forecasting Using Self-Adjusting k-Nearest Neighbours
- 2018
-
In: IET Intelligent Transport Systems. - : Institution of Engineering and Technology. - 1751-956X .- 1751-9578. ; 12:1, s. 41-48
-
Journal article (peer-reviewed)abstract
- Short-term traffic forecasting is becoming more important in intelligent transportation systems. The k-nearest neighbours (kNN) method is widely used for short-term traffic forecasting.However, kNN parameters self-adjustment has been a problem due to dynamic traffic characteristics. This paper proposes a fully automatic dynamic procedure kNN (DP-kNN) that makes the kNN parameters self-adjustable and robust without predefined models or training. We used realworld data with more than one-year traffic records to conduct experiments. The results show that DP-kNN can perform better than manually adjusted kNN and other benchmarking methods with regards to accuracy on average. This study also discusses the difference between holiday and workday traffic prediction as well as the usage of neighbour distance measurement.
|
|
| 10. |
- Wan, Yan, et al.
(author)
-
The orbital effect on the anomalous magnetism and evolution in LaxY1-xVO3 (0 <= x <= 0.2) single crystals
- 2023
-
In: Journal of Alloys and Compounds. - : Elsevier BV. - 0925-8388 .- 1873-4669. ; 932, s. 167526-
-
Journal article (peer-reviewed)abstract
- The orbital effect on the anomalous magnetism and evolution of single crystals with low La doping, LaxY1-xVO3 (x = 0, 0.1, and 0.2), has been studied using single-crystal X-ray diffraction, specific heat, mag-netization, and Raman-scattering techniques. It is found that substituting Y3+ by La3+ increases the de-generacy of the yz/zx orbitals and decreases the Jahn-Teller distortion. These weakens the G-type (antiphase ordering along the c axis) orbital ordering phase. Meanwhile, the substituting decreases the magnetism entropy, indicating the shrinking of the t2g and eg orbital hybridization, eventually destabilizing the C-type (in-phase ordering along the c axis) antiferromagnetic ordering phase. In addition, the mechanism for the shrinking of the diamagnetism with increasing x is analyzed. It may attribute to the competition between the antisymmetric Dzyaloshinsky-Moriya interaction and the single-ion anisotropy.
|
|
