Cluster-based LSTM models to improve Dengue cases forecast
Public health problems such as dengue fever need accurate forecasts so governments can take effective preventive measures. Deep learning (DL) and machine learning have become increasingly popular as the volume of data increases continuously. Nevertheless, performing accurate predictions in areas wi...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Centro Latinoamericano de Estudios en Informática
2023-05-01
|
| Series: | CLEI Electronic Journal |
| Subjects: | |
| Online Access: | https://clei.org/cleiej/index.php/cleiej/article/view/580 |
| _version_ | 1797820111316320256 |
|---|---|
| author | Juan Vicente Bogado Machuca Diego Herbin Stalder Díaz Christian Emilio Schaerer Serra |
| author_facet | Juan Vicente Bogado Machuca Diego Herbin Stalder Díaz Christian Emilio Schaerer Serra |
| author_sort | Juan Vicente Bogado Machuca |
| collection | DOAJ |
| description |
Public health problems such as dengue fever need accurate forecasts so governments can take effective preventive measures. Deep learning (DL) and machine learning have become increasingly popular as the volume of data increases continuously. Nevertheless, performing accurate predictions in areas with fewer cases can be challenging. When we apply DL models using long short-term memory (LSTM) in different cities considering weekly dengue incidence and climate, some models may present heterogeneous behaviours and poor accuracy because of the need for more data. To mitigate this problem, clustering analysis across time series is performed based on scores to measure the clustering quality in 217 Paraguayan cities. First, we compare the raw and feature-based clustering techniques considering several metrics.
Our results indicate that hierarchical clustering combined with Spearman correlation is the most appropriate approach. Finally, several LSTM models built using clustering results were compared. The main contribution of this work is a technique that can improve the performance of time series models that combine information from similar time series and weather data.
|
| first_indexed | 2024-03-13T09:32:32Z |
| format | Article |
| id | doaj.art-92aaff70d144488a94c9d265241f6b05 |
| institution | Directory Open Access Journal |
| issn | 0717-5000 |
| language | English |
| last_indexed | 2024-03-13T09:32:32Z |
| publishDate | 2023-05-01 |
| publisher | Centro Latinoamericano de Estudios en Informática |
| record_format | Article |
| series | CLEI Electronic Journal |
| spelling | doaj.art-92aaff70d144488a94c9d265241f6b052023-05-25T20:38:07ZengCentro Latinoamericano de Estudios en InformáticaCLEI Electronic Journal0717-50002023-05-0126110.19153/cleiej.26.1.4Cluster-based LSTM models to improve Dengue cases forecastJuan Vicente Bogado Machuca0Diego Herbin Stalder Díaz1Christian Emilio Schaerer Serra2National University of Asunción, National University of CaaguazúNational University of Asunción, Engeneering SchoolNational University of Asunción, Politechnic School Public health problems such as dengue fever need accurate forecasts so governments can take effective preventive measures. Deep learning (DL) and machine learning have become increasingly popular as the volume of data increases continuously. Nevertheless, performing accurate predictions in areas with fewer cases can be challenging. When we apply DL models using long short-term memory (LSTM) in different cities considering weekly dengue incidence and climate, some models may present heterogeneous behaviours and poor accuracy because of the need for more data. To mitigate this problem, clustering analysis across time series is performed based on scores to measure the clustering quality in 217 Paraguayan cities. First, we compare the raw and feature-based clustering techniques considering several metrics. Our results indicate that hierarchical clustering combined with Spearman correlation is the most appropriate approach. Finally, several LSTM models built using clustering results were compared. The main contribution of this work is a technique that can improve the performance of time series models that combine information from similar time series and weather data. https://clei.org/cleiej/index.php/cleiej/article/view/580lstmtime series forecastingepidemiologydengue |
| spellingShingle | Juan Vicente Bogado Machuca Diego Herbin Stalder Díaz Christian Emilio Schaerer Serra Cluster-based LSTM models to improve Dengue cases forecast CLEI Electronic Journal lstm time series forecasting epidemiology dengue |
| title | Cluster-based LSTM models to improve Dengue cases forecast |
| title_full | Cluster-based LSTM models to improve Dengue cases forecast |
| title_fullStr | Cluster-based LSTM models to improve Dengue cases forecast |
| title_full_unstemmed | Cluster-based LSTM models to improve Dengue cases forecast |
| title_short | Cluster-based LSTM models to improve Dengue cases forecast |
| title_sort | cluster based lstm models to improve dengue cases forecast |
| topic | lstm time series forecasting epidemiology dengue |
| url | https://clei.org/cleiej/index.php/cleiej/article/view/580 |
| work_keys_str_mv | AT juanvicentebogadomachuca clusterbasedlstmmodelstoimprovedenguecasesforecast AT diegoherbinstalderdiaz clusterbasedlstmmodelstoimprovedenguecasesforecast AT christianemilioschaererserra clusterbasedlstmmodelstoimprovedenguecasesforecast |