Academia.eduAcademia.edu

Figure 12 – uploaded by OGEH CLEMENT

See full PDFdownloadDownload figure
the upper confidence bound or falling below the lower confidence bound. That is, the forecast must fall within the range of the lower confidence bound and the upper confidence bound. or decrease in the prediction but not exceeding Figure 11 shows that Aba in Abia state will continue to experience high rainfall which is likely to result to flooding especially if the control measures ate not put in place to curtail the increase in temperature and also a proactive measute in situations where rainfall increases undermining the high temperature control measutes. The recommendation from this study is that government agencies, ministries, departments and Non-Governmental Organisations (NGOs) concerned should make readily available Geo- data to enable scientist develop Natural Disaster Warning models.

Figure 11 the upper confidence bound or falling below the lower confidence bound. That is, the forecast must fall within the range of the lower confidence bound and the upper confidence bound. or decrease in the prediction but not exceeding Figure 11 shows that Aba in Abia state will continue to experience high rainfall which is likely to result to flooding especially if the control measures ate not put in place to curtail the increase in temperature and also a proactive measute in situations where rainfall increases undermining the high temperature control measutes. The recommendation from this study is that government agencies, ministries, departments and Non-Governmental Organisations (NGOs) concerned should make readily available Geo- data to enable scientist develop Natural Disaster Warning models.

Related Figures (14)

(Maddox, et al., 2002). The historic data can be in years, months, weeks and days divided into sets to construct and evaluate models. To achieve that, the individual set data undergo the training stage of the model, validation, verification and testing stage of the model. The major Machine Language algorithm that have been used for flood prediction are: Artificial Neural Network (ANN), Neuro-Fuzzy Logic, Adaptive Neuro-fuzzy Inference System (ANFIS), Support Vector Machine (SVM), Wavelet Neural Network (WNN) and Multilayered Perceptron (MLP). The chart below describes the flow of activities in creating a Machine Learning Model.
(Maddox, et al., 2002). The historic data can be in years, months, weeks and days divided into sets to construct and evaluate models. To achieve that, the individual set data undergo the training stage of the model, validation, verification and testing stage of the model. The major Machine Language algorithm that have been used for flood prediction are: Artificial Neural Network (ANN), Neuro-Fuzzy Logic, Adaptive Neuro-fuzzy Inference System (ANFIS), Support Vector Machine (SVM), Wavelet Neural Network (WNN) and Multilayered Perceptron (MLP). The chart below describes the flow of activities in creating a Machine Learning Model.
Figure 2. Relationship between Data, Information, Knowledge and Wisdom These data ate useless, unless explored fo. The growth of information technology has resulted to vast amount of data generated and stored in large database and data warehouses. These data are useless, unless explored for
Figure 2. Relationship between Data, Information, Knowledge and Wisdom These data ate useless, unless explored fo. The growth of information technology has resulted to vast amount of data generated and stored in large database and data warehouses. These data are useless, unless explored for
Figure 3. Knowledge Discovery process Source: Nicholas, 2019 Data mining involves processes that start with data as inputs that are processed into information. This is usually called the preprocessing stage that includes variable selections and data cleaning. The information derived is further explored for implicit patterns and rules that were previously unknown. This exploration is done in order to learn from experience the patterns existing in the data. This
Figure 3. Knowledge Discovery process Source: Nicholas, 2019 Data mining involves processes that start with data as inputs that are processed into information. This is usually called the preprocessing stage that includes variable selections and data cleaning. The information derived is further explored for implicit patterns and rules that were previously unknown. This exploration is done in order to learn from experience the patterns existing in the data. This
Table 1. Supervised and Unsupervised Learning
Table 1. Supervised and Unsupervised Learning
This chart is a lower-dimensional projection of the predictor space, which contains a total of 4 predictors.
This chart is a lower-dimensional projection of the predictor space, which contains a total of 4 predictors.
Figure 5. Expert Modeler on Temperature Dataset
Figure 5. Expert Modeler on Temperature Dataset
months. The analysis was completed under 28 microseconds.
months. The analysis was completed under 28 microseconds.
The temperature sequence diagram which completed in 6 seconds, 40 microseconds is thus:
The temperature sequence diagram which completed in 6 seconds, 40 microseconds is thus:
Using the ARIMA (1,1,0) model which ran under 28 microseconds, we have the figure below:
Using the ARIMA (1,1,0) model which ran under 28 microseconds, we have the figure below:
Table 2. Aba Rainfall Forecast Figure 8 depict the classification of data collection centres as classified via the colour of the indications. The indicators with a darker colour are areas that are not flood endangered while those that have lighter colours are flood endangered areas. This is because such areas and centres have high rainfall amongst other contributors to flooding in a place.
Table 2. Aba Rainfall Forecast Figure 8 depict the classification of data collection centres as classified via the colour of the indications. The indicators with a darker colour are areas that are not flood endangered while those that have lighter colours are flood endangered areas. This is because such areas and centres have high rainfall amongst other contributors to flooding in a place.
The classification for rainfall dataset for the data collection centre is given as: Rainfall
The classification for rainfall dataset for the data collection centre is given as: Rainfall
Figure 10. KNN on Rainfall Dataset (Cities) Figure 9. KNN on Rainfall Dataset (States)
Figure 10. KNN on Rainfall Dataset (Cities) Figure 9. KNN on Rainfall Dataset (States)
The classification for rainfall dataset for the states is given as:
The classification for rainfall dataset for the states is given as:
Related topics:
GeographyArchaeologyComputer ScienceArtificial IntelligenceMachine LearningBiologyEcology K-nearest neighbors algorithmStatistical ClassificationFlood Myth
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved