PREDICTION OF HYPERTENSION RISKS WITH FEATURE SELECTION AND XGBOOST

There are about 1 billion hypertensives patients on a global scale. Hypertension has become the main cause of shorter lifespan and disability for humans worldwide. In this essay, we constructed a new model based on hybrid feature selection and the standard XGBoost for hypertension detection and prediction. After having successfully utilized Lasso regression to identify hypertension-related factors, we used the standard XGBoost model for hypertension prediction. The result from the experiments conducted on the data from the BRFSS shows that proposed model can achieve 77.2% accuracy and 84.6% AUC, both about 7% higher than that without the nonoptimized model. Our proposed model can not only be used to predict the risk of hypertension, but also provide customers with suggestions on how to lead a healthy lifestyle.

Download Full-text

A Regional Zenith Tropospheric Correction Model Based on the Guangxi CORS Network

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.353-356.3438 ◽

2013 ◽

Vol 353-356 ◽

pp. 3438-3443

Author(s):

Li Long Liu ◽

Liang Ke Huang ◽

Teng Xu Zhang ◽

Miao Zhou ◽

Chao Long Yao

Keyword(s):

Interpolation Method ◽

Meteorological Data ◽

Correction Model ◽

New Model ◽

Model Based ◽

Proposed Model ◽

New Models ◽

Using Data ◽

The Relationship ◽

Tropospheric Delays

In this paper, the relationship between zenith tropospheric delays and the altitude of stations is analyzed using the EGNOS tropospheric correction model. The new model (EHT model) is proposed for estimating zenith tropospheric delays from regional CORS data without meteorological data. The proposed model is compared with the direct interpolation method and the remove-restore method using data from Guangxi CORS. The results show that the new models significantly improve the calculated precision.

Download Full-text

A Combined Model Based on Feature Selection and WOA for PM2.5 Concentration Forecasting

Atmosphere ◽

10.3390/atmos10040223 ◽

2019 ◽

Vol 10 (4) ◽

pp. 223 ◽

Cited By ~ 7

Author(s):

Fang Zhao ◽

Weide Li

Keyword(s):

Feature Selection ◽

Correlation Coefficients ◽

Data Preprocessing ◽

Parameters Optimization ◽

General Regression Neural Network ◽

Support Vector ◽

Model Based ◽

Mode Decomposition ◽

Proposed Model ◽

Whale Optimization

As people pay more attention to the environment and health, P M 2.5 receives more and more consideration. Establishing a high-precision P M 2.5 concentration prediction model is of great significance for air pollutants monitoring and controlling. This paper proposed a hybrid model based on feature selection and whale optimization algorithm (WOA) for the prediction of P M 2.5 concentration. The proposed model included five modules: data preprocessing module, feature selection module, optimization module, forecasting module and evaluation module. Firstly, signal processing technology CEEMDAN-VMD (Complete Ensemble Empirical Mode Decomposition with Adaptive Noise and Variational Mode Decomposition) is used to decompose, reconstruct, identify and select the main features of P M 2.5 concentration series in data preprocessing module. Then, AutoCorrelation Function (ACF) is used to extract the variables which have relatively large correlation with predictor, so as to select input variables according to the order of correlation coefficients. Finally, Least Squares Support Vector Machine (LSSVM) is applied to predict the hourly P M 2.5 concentration, and the parameters of LSSVM are optimized by WOA. Two experiment studies reveal that the performance of the proposed model is better than benchmark models, such as single LSSVM model with default parameters optimization, single BP neural networks (BPNN), general regression neural network (GRNN) and some other combined models recently reported.

Download Full-text

A New Model Based on Artificial Neural Networks and Game Theory for the Selection of Underground Mining Method

Физико-технические проблемы разработки полезных ископаемых ◽

10.15372/ftprpi20200108 ◽

2020 ◽

Keyword(s):

Neural Networks ◽

Game Theory ◽

Artificial Neural Networks ◽

Underground Mining ◽

Mining Method ◽

New Model ◽

Model Based ◽

Artificial Neural ◽

Selection Of

Download Full-text

A Modified Dugoff Tire Model for Combined-slip Forces

Tire Science and Technology ◽

10.2346/1.3481696 ◽

2010 ◽

Vol 38 (3) ◽

pp. 228-244 ◽

Cited By ~ 19

Author(s):

Nenggen Ding ◽

Saied Taheri

Keyword(s):

Vehicle Dynamics ◽

Tire Model ◽

Magic Formula ◽

Model Based ◽

Proposed Model ◽

Road Friction ◽

On Line ◽

Double Lane Change ◽

Tire Forces ◽

Tire Models

Abstract Easy-to-use tire models for vehicle dynamics have been persistently studied for such applications as control design and model-based on-line estimation. This paper proposes a modified combined-slip tire model based on Dugoff tire. The proposed model takes emphasis on less time consumption for calculation and uses a minimum set of parameters to express tire forces. Modification of Dugoff tire model is made on two aspects: one is taking different tire/road friction coefficients for different magnitudes of slip and the other is employing the concept of friction ellipse. The proposed model is evaluated by comparison with the LuGre tire model. Although there are some discrepancies between the two models, the proposed combined-slip model is generally acceptable due to its simplicity and easiness to use. Extracting parameters from the coefficients of a Magic Formula tire model based on measured tire data, the proposed model is further evaluated by conducting a double lane change maneuver, and simulation results show that the trajectory using the proposed tire model is closer to that using the Magic Formula tire model than Dugoff tire model.

Download Full-text

Applied Design Structure Matrices to Quantify Impact of Proposed Model Based Engineering Solutions

AIAA Scitech 2021 Forum ◽

10.2514/6.2021-0238 ◽

2021 ◽

Author(s):

Shana West ◽

Matthew W. Cribb ◽

Thomas R. McCarthy

Keyword(s):

Model Based ◽

Design Structure ◽

Proposed Model ◽

Applied Design

Download Full-text

A new path loss model based on the volumetric occupancy rate for the pine forests at 5G frequency band

International Journal of Microwave and Wireless Technologies ◽

10.1017/s175907872000152x ◽

2020 ◽

pp. 1-10

Author(s):

Abdullah Genc

Keyword(s):

Path Loss ◽

Forest Area ◽

Occupancy Rate ◽

Loss Model ◽

Path Loss Model ◽

Model Based ◽

Proposed Model ◽

Pine Trees ◽

Foliage Density ◽

Reference Measurements

Abstract In this paper, a new empirical path loss model based on frequency, distance, and volumetric occupancy rate is generated at the 3.5 and 4.2 GHz in the scope of 5G frequency bands. This study aims to determine the effect of the volumetric occupancy rate on path loss depending on the foliage density of the trees in the pine forest area. Using 4.2 GHz and the effect of the volumetric occupancy rate contributes to the literature in terms of novelty. Both the reference measurements to generate a model and verification measurements to verify the proposed models are conducted in three different regions of the forest area with double ridged horn antennas. These regions of the artificial forest area consist of regularly sorted and identical pine trees. Root mean square error (RMSE) and R-squared values are calculated to evaluate the performance of the proposed model. For 3.5 and 4.2 GHz, while the RMSEs are 3.983 and 3.883, the values of R-squared are 0.967 and 0.963, respectively. Additionally, the results are compared with four path loss models which are commonly used in the forest area. The proposed one has the best performance among the other models with values 3.98 and 3.88 dB for 3.5 and 4.2 GHz.

Download Full-text

Lane Departure Warning Estimation Using Yaw Acceleration

Open Engineering ◽

10.1515/eng-2021-0008 ◽

2020 ◽

Vol 11 (1) ◽

pp. 102-111

Author(s):

Em Poh Ping ◽

J. Hossen ◽

Wong Eng Kiong

Keyword(s):

Vehicle Dynamics ◽

Dynamic Responses ◽

Dynamics Simulation ◽

Steering Wheel ◽

Mathematical Representation ◽

Vehicle Speed ◽

Lane Departure Warning ◽

Model Based ◽

Lane Departure ◽

Proposed Model

AbstractLane departure collisions have contributed to the traffic accidents that cause millions of injuries and tens of thousands of casualties per year worldwide. Due to vision-based lane departure warning limitation from environmental conditions that affecting system performance, a model-based vehicle dynamics framework is proposed for estimating the lane departure event by using vehicle dynamics responses. The model-based vehicle dynamics framework mainly consists of a mathematical representation of 9-degree of freedom system, which permitted to pitch, roll, and yaw as well as to move in lateral and longitudinal directions with each tire allowed to rotate on its axle axis. The proposed model-based vehicle dynamics framework is created with a ride model, Calspan tire model, handling model, slip angle, and longitudinal slip subsystems. The vehicle speed and steering wheel angle datasets are used as the input in vehicle dynamics simulation for predicting lane departure event. Among the simulated vehicle dynamic responses, the yaw acceleration response is observed to provide earlier insight in predicting the future lane departure event compared to other vehicle dynamics responses. The proposed model-based vehicle dynamics framework had shown the effectiveness in estimating lane departure using steering wheel angle and vehicle speed inputs.

Download Full-text

RoRePo: Detecting the role information and relative position information for contexts in multi-turn dialogue generation

Journal of Intelligent & Fuzzy Systems ◽

10.3233/jifs-202641 ◽

2021 ◽

Vol 40 (5) ◽

pp. 10003-10015

Author(s):

Zibang Gan ◽

Biqing Zeng ◽

Lianglun Cheng ◽

Shuai Liu ◽

Heng Yang ◽

...

Keyword(s):

Relative Position ◽

Latent Variables ◽

Role Ambiguity ◽

Position Information ◽

New Model ◽

Proposed Model ◽

Relevant Context ◽

Position Representation ◽

Current Speaker ◽

Relative Position Information

In multi-turn dialogue generation, dialogue contexts have been shown to have an important influence on the reasoning of the next round of dialogue. A multi-turn dialogue between two people should be able to give a reasonable response according to the relevant context. However, the widely used hierarchical recurrent encoder-decoder model and the latest model that detecting the relevant contexts with self-attention are facing the same problem. Their given response doesn’t match the identity of the current speaker, which we call it role ambiguity. In this paper, we propose a new model, named RoRePo, to tackle this problem by detecting the role information and relative position information. Firstly, as a part of the decoder input, we add a role embedding to identity different speakers. Secondly, we incorporate self-attention mechanism with relative position representation to dialogue context understanding. Besides, the design of our model architecture considers the influence of latent variables in generating more diverse responses. Experimental results of our evaluations on the DailyDialog and DSTC7_AVSD datasets show that our proposed model advances in multi-turn dialogue generation.

Download Full-text