A Discussion of Data Sampling Strategies for Early Action Prediction

2022 ◽  
pp. 306-314
Author(s):  
Xiaofa Liu ◽  
Xiaoli Liu ◽  
Jianqin Yin
Keyword(s):  
Sampling Strategies ◽  
Data Sampling ◽  
Action Prediction ◽  
Early Action

Early Action Prediction Using 3DCNN With LSTM and Bidirectional LSTM

2021 ◽  
Author(s):  
Manju D ◽  
Dr. Seetha M. ◽  
Dr. Sammulal P.
Keyword(s):  
Action Prediction ◽  
Early Action ◽  
Bidirectional Lstm

scholarly journals A Multivariate Empirical Orthogonal Function Method to Construct Nitrate Maps in the Southern Ocean

2018 ◽  
Vol 35 (7) ◽  
pp. 1505-1519 ◽  
Author(s):  
Yu-Chiao Liang ◽  
Matthew R. Mazloff ◽  
Isabella Rosso ◽  
Shih-Wei Fang ◽  
Jin-Yi Yu
Keyword(s):  
Southern Ocean ◽  
General Circulation ◽  
Model Simulation ◽  
Potential Temperature ◽  
Circulation Model ◽  
Empirical Orthogonal Functions ◽  
Cluster Method ◽  
Sampling Strategies ◽  
Data Sampling ◽  
Mean Square Errors

AbstractThe ability to construct nitrate maps in the Southern Ocean (SO) from sparse observations is important for marine biogeochemistry research, as it offers a geographical estimate of biological productivity. The goal of this study is to infer the skill of constructed SO nitrate maps using varying data sampling strategies. The mapping method uses multivariate empirical orthogonal functions (MEOFs) constructed from nitrate, salinity, and potential temperature (N-S-T) fields from a biogeochemical general circulation model simulation Synthetic N-S-T datasets are created by sampling modeled N-S-T fields in specific regions, determined either by random selection or by selecting regions over a certain threshold of nitrate temporal variances. The first 500 MEOF modes, determined by their capability to reconstruct the original N-S-T fields, are projected onto these synthetic N-S-T data to construct time-varying nitrate maps. Normalized root-mean-square errors (NRMSEs) are calculated between the constructed nitrate maps and the original modeled fields for different sampling strategies. The sampling strategy according to nitrate variances is shown to yield maps with lower NRMSEs than mapping adopting random sampling. A k-means cluster method that considers the N-S-T combined variances to identify key regions to insert data is most effective in reducing the mapping errors. These findings are further quantified by a series of mapping error analyses that also address the significance of data sampling density. The results provide a sampling framework to prioritize the deployment of biogeochemical Argo floats for constructing nitrate maps.

scholarly journals EARLY ACTION PREDICTION USING VGG16 MODEL AND BIDIRECTIONAL LSTM

2021 ◽  
Vol 9 (1) ◽  
pp. 666-672
Author(s):  
Manju D, Dr. Seetha M, Dr. Sammulal P
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Video Surveillance ◽  
Early Stage ◽  
Early Prediction ◽  
Action Prediction ◽  
Automated Driving ◽  
Time Step ◽  
Driving System ◽  
Early Action

Action prediction plays a key function, where an expected action needs to be identified before the action is completely performed. Prediction means inferring a potential action until it occurs at its early stage. This paper emphasizes on early action prediction, to predict an action before it occurs. In real time scenarios, the early prediction can be very crucial and has many applications like automated driving system, healthcare, video surveillance and other scenarios where a proactive action is needed before the situation goes out of control. VGG16 model is used for the early action prediction which is a convolutional neural network with 16 layers depth. Besides its capability of classifying objects in the frames, the availability of model weights enhances its capability. The model weights are available freely and preferred to used in different applications or models. The VGG-16 model along with Bidirectional structure of Lstm enables the network to provide both backward and forward information at every time step. The results of the proposed approach increased observation ratio ranging from 0.1 to 1.0 compared with the accuracy of GAN model.

scholarly journals Early Action Prediction With Generative Adversarial Networks

IEEE Access ◽  
2019 ◽  
Vol 7 ◽  
pp. 35795-35804 ◽  
Author(s):  
Dong Wang ◽  
Yuan Yuan ◽  
Qi Wang
Keyword(s):  
Generative Adversarial Networks ◽  
Action Prediction ◽  
Adversarial Networks ◽  
Early Action

Statistical Modeling for Ceramic Analysis

2016 ◽  
pp. 57-72
Author(s):  
Gulsebnem Bishop
Keyword(s):  
Complex Data ◽  
Sampling Strategies ◽  
Data Sampling ◽  
Ceramic Analysis ◽  
Chi Square ◽  
Statistical Population ◽  
Archaeological Data ◽  
Principle Components Analysis ◽  
Chi Square Test ◽  
Components Analysis

Statistics can be used to describe, model, and predict archaeological data, provided that the analyst has an understanding of the strengths and limitations of their data type and has a well-defined statistical population. This chapter discusses the major types of archaeological data, sampling strategies, and statistics appropriate for both describing and predicting outcomes for simple and complex ceramic datasets. Description and modeling of complex data can be done with many tools ranging from simple charts and histograms to more complicated methods, such as T-Test, Chi-Square Test, Multi-Response Permutation Procedure (MRPP), and Kernel Density Estimation (KDE), as well as Principle Components Analysis (PCA).

Analyzing Internet Forums

2012 ◽  
Vol 24 (2) ◽  
pp. 55-66 ◽  
Author(s):  
Peter Holtz ◽  
Nicole Kronberger ◽  
Wolfgang Wagner
Keyword(s):  
Discussion Board ◽  
Computer Assisted ◽  
Sampling Strategies ◽  
Data Sampling ◽  
Online Forums ◽  
New Members ◽  
Depth Analysis ◽  
Internet Forums ◽  
Social Scientific ◽  
Matters Of Concern

Within Internet forums, members of certain (online) communities discuss matters of concern to the respective groups, with comparatively few social restraints. For radical, extremist, and other ideologically “sensitive” groups and organizations in particular, Internet forums are a very efficient and widely used tool to connect members, inform others about the group’s agenda, and attract new members. Whereas members of such groups may be reluctant to express their opinions in interviews or surveys, we argue that Internet forums can yield an abundance of useful “natural” discursive data for social scientific research. Based on two exemplary studies, we present a practical guide for the analysis of such data, including data-sampling strategies, the refinement of the data for computer-assisted qualitative and quantitative analysis, and strategies for in-depth analysis. The first study is an in-depth analysis of discourses within a German neo-Nazi discussion board. In the second, nine online forums for young German Muslims were analyzed and compared. Advantages and potential issues with analyzing Internet forums are discussed.

Give Chance a Chance: Modeling Density to Enhance Scatter Plot Quality through Random Data Sampling

2006 ◽  
Vol 5 (2) ◽  
pp. 95-110 ◽  
Author(s):  
Enrico Bertini ◽  
Giuseppe Santucci
Keyword(s):  
Information Visualization ◽  
Quality Metrics ◽  
Virtual Space ◽  
Sampling Strategies ◽  
Data Sampling ◽  
Feature Preservation ◽  
Formal Framework ◽  
Absolute Density ◽  
Automatic Sampling ◽  
Scatter Plots

The problem of visualizing huge amounts of data is well known in information visualization. Dealing with a large number of items forces almost any kind of Infovis technique to reveal its limits in terms of expressivity and scalability. In this paper we focus on 2D scatter plots, proposing a ‘feature preservation’ approach, based on the idea of modeling the visualization in a virtual space in order to analyze its features (e.g., absolute density, relative density, etc.). In this way we provide a formal framework to measure the visual overlapping, obtaining precise quality metrics about the visualization degradation and devising automatic sampling strategies able to improve the overall image quality. Metrics and algorithms have been improved through suitable user studies.

Sign in / Sign up

Export Citation Format

Share Document