scholarly journals A Practical and Adaptive Approach to Predicting Indoor CO2

2021 ◽  
Vol 11 (22) ◽  
pp. 10771
Author(s):  
Giacomo Segala ◽  
Roberto Doriguzzi-Corin ◽  
Claudio Peroni ◽  
Tommaso Gazzini ◽  
Domenico Siracusa
Keyword(s):  
Data Collection ◽  
Prediction Accuracy ◽  
Network Models ◽  
Window Size ◽  
Environmental Data ◽  
Neural Network Models ◽  
Using Data ◽  
Carbon Dioxide Co2 ◽  
Long Term Performance ◽  
Term Performance

COVID-19 has underlined the importance of monitoring indoor air quality (IAQ) to guarantee safe conditions in enclosed environments. Due to its strict correlation with human presence, carbon dioxide (CO2) represents one of the pollutants that most affects environmental health. Therefore, forecasting future indoor CO2 plays a central role in taking preventive measures to keep CO2 level as low as possible. Unlike other research that aims to maximize the prediction accuracy, typically using data collected over many days, in this work we propose a practical approach for predicting indoor CO2 using a limited window of recent environmental data (i.e., temperature; humidity; CO2 of, e.g., a room, office or shop) for training neural network models, without the need for any kind of model pre-training. After just a week of data collection, the error of predictions was around 15 parts per million (ppm), which should enable the system to regulate heating, ventilation and air conditioning (HVAC) systems accurately. After a month of data we reduced the error to about 10 ppm, thereby achieving a high prediction accuracy in a short time from the beginning of the data collection. Once the desired mobile window size is reached, the model can be continuously updated by sliding the window over time, in order to guarantee long-term performance.

scholarly journals Gestation at completion of prenatal questionnaires in ALSPAC

2020 ◽  
Vol 5 ◽  
pp. 100
Author(s):  
Yasmin Iles-Caven ◽  
Kate Northstone ◽  
Jean Golding
Keyword(s):  
Data Collection ◽  
Pregnant Women ◽  
First Trimester ◽  
Environmental Data ◽  
Parents And Children ◽  
Using Data ◽  
Time Critical ◽  
Avon Longitudinal Study ◽  
In Pregnancy ◽  
Made In

Enrolling a cohort in pregnancy can be methodologically difficult in terms of structuring data collection. For example, some exposures of interest may be time-critical while other (often retrospective) data can be collected at any point during pregnancy.  The Avon Longitudinal Study of Parents and Children (ALSPAC) is a prime example of a cohort where certain data were collected at specific time points and others at variable times depending on the gestation at contact.  ALSPAC aimed to enrol as many pregnant women as possible in a geographically defined area with an expected date of delivery between April 1991 and December 1992. The ideal was to enrol women as early in pregnancy as possible, and to collect information, when possible, at two fixed gestational periods (18 and 32 weeks). A variety of methods were used to enrol participants.   Approximately 80% of eligible women resident in the study area were enrolled. Gestation at enrolment ranged from 4-41 (median = 14) weeks of pregnancy. Given this variation in gestation we describe the various decisions that were made in regard to the timing of questionnaires to ensure that appropriate data were obtained from the pregnant women.  45% of women provided data during the first trimester, this is less than ideal but reflects the fact that many women do not acknowledge their pregnancy until the first trimester is safely completed. Data collection from women at specific gestations (18 and 32 weeks) was much more successful (80-85%). Unfortunately, it was difficult to obtain environmental data during the first trimester. Given the time critical nature of exposures during this trimester, researchers must take the gestational age at which environmental data was collected into account. This is particularly important for data collected using the questionnaire named ‘Your Environment’ (using data known as the A files).

Neural Network-Based Stock Market Return Forecasting Using Data Mining for Variable Reduction

2008 ◽  
pp. 2476-2493 ◽  
Author(s):  
David Encke
Keyword(s):  
Neural Network ◽  
Information Gain ◽  
Network Models ◽  
Linear Regression Models ◽  
Market Returns ◽  
Neural Network Models ◽  
The Neural Network ◽  
Variable Reduction ◽  
Input Variables ◽  
Using Data

Researchers have known for some time that nonlinearity exists in the financial markets and that neural networks can be used to forecast market returns. Unfortunately, many of these studies fail to consider alternative forecasting techniques, or the relevance of the input variables. The following research utilizes an information-gain technique from machine learning to evaluate the predictive relationships of numerous financial and economic input variables. Neural network models for level estimation and classification are then examined for their ability to provide an effective forecast of future values. A cross-validation technique is also employed to improve the generalization ability of the models. The results show that the classification models generate higher accuracy in forecasting ability than the buy-and-hold strategy, as well as those guided by the level-estimation-based forecasts of the neural network and benchmark linear regression models.

Lifetimes of Titanium and Copper Containers for the Disposal of Used Nuclear Fuel

1991 ◽  
Vol 257 ◽  
Author(s):  
Lawrence H. Johnson ◽  
D.W. Shoesmith ◽  
B.M. Ikeda ◽  
F. King
Keyword(s):  
Nuclear Fuel ◽  
Bulk Solution ◽  
General Corrosion ◽  
Compacted Clay ◽  
Used Nuclear Fuel ◽  
Buffer Material ◽  
Using Data ◽  
Long Term Performance ◽  
Term Performance ◽  
Nuclear Fuel Waste

ABSTRACTTitanium and copper have been proposed as suitable container materials for disposal of nuclear fuel waste in plutonic rock of the Canadian Shield. Studies of the corrosion of these materials have led to the development of container failure models to predict long-term performance. Crevice corrosion and hydrogen-induced cracking of titanium have been identified as potential failure mechanisms, and these two processes have been studied in detail. Using data from these studies as well as a number of conservative assumptions, titanium container lifetimes of 1200 to 7000 a have been estimated. For copper, general corrosion has been studied in detail in bulk solution and in compacted clay-based buffer material. Results indicate that the copper corrosion rate is likely to be controlled by the rate of transport of copper species away from the container surface. An assessment of copper pitting data suggests that pitting is an extremely improbable failure mechanism. The copper container failure model predicts minimum container lifetimes of 30 000 a. The results demonstrate that long lifetime containment can be provided, should performance assessment studies indicate the need for such an option.

scholarly journals Design and Implementation of a Statistical Testing Framework for a Lightweight Stream Cipher

2020 ◽  
Vol 10 (1) ◽  
pp. 5132-5141
Author(s):  
A. Alamer ◽  
B. Soh
Keyword(s):  
Stream Cipher ◽  
Network Models ◽  
Window Size ◽  
Binary Sequences ◽  
Statistical Testing ◽  
Linear Feedback ◽  
Maximal Degree ◽  
Neural Network Models ◽  
Computing Power ◽  
Testing Framework

The Shrinking Generator (SG) is a popular synchronous, lightweight stream cipher that uses minimal computing power. However, its strengths and weaknesses have not been studied in detail. This paper proposes a statistical testing framework to assess attacks on the SG. The framework consists of a d-monomial test that is adapted to SG by applying the algebraic normal form (ANF) representation of Boolean functions, a test that uses the maximal degree monomial test to determine whether the ANF follows the proper mixing of bit values, and a proposed unique window size (UWS) scheme to test the randomness properties of the keystream. The proposed framework shows significant weaknesses in the SG output in terms of dependence between the controlling linear-feedback shift register (LFSR) and non-linearity of the resulting keystream. The maximal degree monomial test provides a better understanding of the optimal points of SG, demonstrating when it is at its best and worst according to the first couple of results. This paper uses UWS to illustrate the effect of the LFSR choice on possibly distinguishing attacks on the SG. The results confirm that the proposed UWS scheme is a viable measure of the cryptographic strength of a stream cipher. Due to the importance of predictability and effective tools, we used neural network models to simulate the input data for the pseudo-random binary sequences. Through the calculation of UWS, we obtained solid results for the predictions.

scholarly journals Gated Graph Attention Network for Cancer Prediction

Sensors ◽  
2021 ◽  
Vol 21 (6) ◽  
pp. 1938
Author(s):  
Linling Qiu ◽  
Han Li ◽  
Meihong Wang ◽  
Xiaoli Wang
Keyword(s):  
Neural Network ◽  
Prediction Accuracy ◽  
State Of The Art ◽  
Network Models ◽  
The State ◽  
Neural Network Models ◽  
Attention Network ◽  
Training Time ◽  
Cancer Prediction ◽  
Gating Mechanism

With its increasing incidence, cancer has become one of the main causes of worldwide mortality. In this work, we mainly propose a novel attention-based neural network model named Gated Graph ATtention network (GGAT) for cancer prediction, where a gating mechanism (GM) is introduced to work with the attention mechanism (AM), to break through the previous work’s limitation of 1-hop neighbourhood reasoning. In this way, our GGAT is capable of fully mining the potential correlation between related samples, helping for improving the cancer prediction accuracy. Additionally, to simplify the datasets, we propose a hybrid feature selection algorithm to strictly select gene features, which significantly reduces training time without affecting prediction accuracy. To the best of our knowledge, our proposed GGAT achieves the state-of-the-art results in cancer prediction task on LIHC, LUAD, KIRC compared to other traditional machine learning methods and neural network models, and improves the accuracy by 1% to 2% on Cora dataset, compared to the state-of-the-art graph neural network methods.

Neural Network-Based Stock Market Return Forecasting Using Data Mining for Variable Reduction

2006 ◽  
pp. 43-63
Author(s):  
David Encke
Keyword(s):  
Neural Network ◽  
Information Gain ◽  
Network Models ◽  
Linear Regression Models ◽  
Market Returns ◽  
Neural Network Models ◽  
The Neural Network ◽  
Variable Reduction ◽  
Input Variables ◽  
Using Data

Researchers have known for some time that nonlinearity exists in the financial markets and that neural networks can be used to forecast market returns. Unfortunately, many of these studies fail to consider alternative forecasting techniques, or the relevance of the input variables. The following research utilizes an information-gain technique from machine learning to evaluate the predictive relationships of numerous financial and economic input variables. Neural network models for level estimation and classification are then examined for their ability to provide an effective forecast of future values. A cross-validation technique is also employed to improve the generalization ability of the models. The results show that the classification models generate higher accuracy in forecasting ability than the buy-and-hold strategy, as well as those guided by the level-estimation-based forecasts of the neural network and benchmark linear regression models.

scholarly journals Predictive Ability of Improved Neural Network Models to Simulate Pollutant Dispersion

2014 ◽  
Vol 2014 ◽  
pp. 1-12 ◽  
Author(s):  
Khandaker M. A. Hossain
Keyword(s):  
Neural Network ◽  
Predictive Ability ◽  
Network Models ◽  
Pollutant Dispersion ◽  
Neural Network Models ◽  
Coastal Cities ◽  
Ann Models ◽  
Using Data ◽  
Artificial Neural Network Ann ◽  
Urban Atmospheres

This paper describes the ability of artificial neural network (ANN) models to simulate the pollutant dispersion characteristics in varying urban atmospheres at different regions. ANN models are developed based on twelve meteorological (including rainfall/precipitation) and six traffic parameters/variables that have significant influence on emission/pollutant dispersion. The models are trained to predict concentration of carbon monoxide and particulate matters in urban atmospheres using field meteorological and traffic data. Training, validation, and testing of ANN models are conducted using data from the Dhaka city of Bangladesh. The models are used to simulate concentration of pollutants as well as the effect of rainfall on emission dispersion throughout the year and inversion condition during the night. The predicting ability and robustness of the models are then determined by using data of the coastal cities of Chittagong and Dhaka. ANN models based on both meteorological and traffic variables exhibit the best performance and are capable of resolving patterns of pollutant dispersion to the atmosphere for different cities.

Sign in / Sign up

Export Citation Format

Share Document