Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 7

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  linear regression analysis
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
EN
This paper proposes an enhancement approach to improve the accuracy of global Digital Elevation Models (GDEMs) in Egypt. The proposed approach is an empirical one that depends on subtracting the heights error from the original DEM. The research includes the evaluation and enhancement of SRTM-1 (SRTM v4.1), ASTER GDEM v2, and AW3D30 v2 GDEMs, in Egypt, using 980 well distributed GPS/levelling points, that cover the entire country. The GPS/levelling points are divided into 500 control and 390 check points. The results show that the root mean square error (RMSE) in the SRTM, ASTER and AW3D30 are 3.99 m, 8.81 m, and 2.98 m respectively. For enhancing purposes, two different approaches are used: a linear regression analysis approach, and the proposed empirical surface subtraction approach. The results of the linear regression analysis approach show that the accuracies are improved by 3%, 16%, and 3% for SRTM, ASTER and AW3D30 respectively. However, the accuracies are improved by 5%, 23%, and 16% for SRTM, ASTER and AW3D30 respectively when the proposed approach is followed. After using the proposed approach, the obtained accuracy of the enhanced DEM reached 2.5 m.
EN
One of the most challenging issues regarding water quality control is the lack of adequate measurements and lack of data on many water quality constituents. Since water quality is highly variable during time and space, traditional grab sampling often misses extreme events and the result isn’t always a representative one. This paper evaluated the usefulness of turbidity measurement as a surrogate for the determination of the Total Phosphorus. Instead of in-situ sensors for real time turbidity measurements, another approach was taken during this investigation. The suitability of using turbidity as a surrogate for TP was investigated using prepared subsamples, each with different concentration of water quality constituents. Laboratory results, with the use of the linear regression techniques, enabled the development of the model that relates the total phosphorus to turbidity. The linear regression equation developed, TP = 0.3942TTU – 3.4279 shows that there is a very good prediction of the total phosphorus based on the turbidity measurement, with the correlation coefficient as good as R2 = 0.8782 and p-value less than 0.5. Even though the equation is site specific, and more investigation is needed, we conclude that it can be used in similar situations, when there is a lack of monitoring programs.
EN
In construction practice, contractually agreed costs are often exceeded, which interferes with the sustainable realization of construction projects. The research described in this paper covers 24 new construction, renovation and reconstruction projects in the Republic of Croatia realized in the years 2006 to 2017, in order to analyse the occurrence of cost overruns more precisely with regard to the source of the overruns. It was found that additional work is the main source of cost overruns: firstly, additional work as a result of the client’s change orders and then unforeseen construction work as a result of unforeseen circumstances. As for the additional works, they are carried out at the client’s request and are not necessary for the safety and stability of the building. Using linear regression and “soft computing” methods, the possibility of modelling the relationship between contractually agreed and realized construction costs with satisfactory accuracy was tested. The model with the values of the natural logarithms of the variables, modelled according to the time–cost model of Bromilow, proved to be of the highest accuracy.
PL
W ostatnich latach tworzenie obywatelskich (społecznościowych) danych przestrzennych przez użytkowników Internetu, niebędących profesjonalistami w tym zakresie, jest coraz bardziej popularne. Świadczy o tym również rosnąca liczba inicjatyw opartych o dane zbierane na zasadzie crowdsourcingu (ang. crowd – tłum, ang. sourcing – czerpanie). Przyczynia się to do wzrostu świadomości społecznej dotyczącej danych geoprzestrzennych. Celem artykułu było zbadanie jakie cechy społeczeństwa wpływają na zaangażowanie obywateli w tworzenie VGI (ang. volunteered geographic information) w Polsce. Do jego realizacji wykorzystano dane z projektu OpenStreetMap oraz dane charakteryzujące społeczeństwo pozyskane z Głównego Urzędu Statystycznego. Były to między innymi: poziom wykształcenia, miesięczne wynagrodzenie, współczynnik feminizacji. Pierwszym etapem było określenie stopnia korelacji między danymi opisującymi społeczeństwo a danymi pozyskanymi w projekcie OpenStreetMap w podziale na powiaty. Następnie dla najbardziej skorelowanych zmiennych ułożono modele regresji wielorakiej i regresji ważonej geograficznie (GWR), co pozwoliło na wyznaczenie tych cech społeczeństwa, które miały istotny wpływ na pozyskiwanie VGI w Polsce.
EN
In recent years, the creation of volunteered geographic information (VGI) by Internet users, who are not professionals in this area is becoming increasingly popular. There is also a growing number of initiatives based on the data collected on the basis of crowdsourcing. This contributes to increase of the public awareness of geospatial data. The aim of the paper was to examine what features of socjety affect the involvement of citizens in creating VGI in Poland. To achieve this objective, data from the OpenStreetMap project and society data obtained from the Central Statistical Office (this included level of education, monthly salary, the feminisation rate) were used. The first stage was to determine the degree of correlation between the data describing the society, and the OpenStreetMap data divided into districts. Then, for the most correlated variables multiple regression and geographically weighted regression (GWR) models were arranged, which allowed the determination of the characteristics of a society that had a significant effect on the acquisition of VGI in Poland.
EN
An ideal broiler house should be designed to minimize the effects of weather changes and to keep indoor conditions at the comfort temperature of the animals. In this case, this should be done with minimum cost and possible lowest operating costs. Degree-day methods are used in order to have knowledge of the energy need of any structure. With this method, the measured values or meteorological data can be used to give information about the heating and cooling energy quantities of structures. Depending on the climate change in recent years, the changes need to be examined that have taken place in order to provide optimum comfort in animal barn. Isparta province and districts were selected as the study area. The longterm average daily temperature values are used from meteorological stations of the selected region. The heating and cooling degree day values were calculated for selected balance temperatures in broiler breeding. Linear Regression Analysis and Spearman Rank Correlation Test were conducted to determine the changes of these values due to climate change. In conclusion, it was determined that there were statistically significant trends at 5% significance level in Egirdir (21°C), Isparta (31-29-25 and 23°C), Kasimlar (18°C), Senirkent (31-29-25-23 and 21°C), Sutculer (all selected balance temperature values) and Yalvac (31-29-25-23 and 21°C) in terms of heating degree-day values, and in Atabey (29-25-23-21 and 18°C), Barla (25-23 and 21°C), Isparta (23-21 and 18°C), Senirkent (29- 25-23-21 and 18°C), Sutculer (29-25-23-21 and 18°C), Sarkikaraagac (25-23-21 and 18°C), Uluborlu (25-23-21 and 18°C) and Yalvac (25-23-21 and 18°C) in terms of cooling degree-day values. As a result, it has been concluded that more energy consumption will be a concern for heating and cooling of the broiler house that will be built in the province of Isparta.
EN
The paper deals with the recent survey of indoor radon (Rn) results in schools, where paired CR-39 detectors were simultaneously exposed to different long-term periods, i.e., one detector was exposed during the whole year and the other one in the period of the school year duration. To be able to compare the results obtained, for its analysis, the relative bias and U tests were used. It was found that there are no systematic differences between the results, which points that the exposure of the detector during summer vacations did not affect the estimated average annual radon concentration. The paired results were modelled by a linear function, giving an extremely high coeffi cient of determination R2 = 0.99.
EN
The determinants method of explanatory variables set selection to the linear model is shown in this article. This method is very useful to find such a set of variables which satisfy small relative error of the linear model as well as small relative error of parameters estimation of this model. Knowledge of the values of the parameters of this model is not necessary. An example of the use of the determinants method for world’s population model is also shown in this article. This method was tested for 224 – 1 models for a set of 23 potential explanatory variables. 5 world’s population models with one, two, three, four and five explanatory variables were chosen and analysed.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.