发明名称 |
Forecasting hotspots using predictive visual analytics approach |
摘要 |
A method for forecasting hotspots is provided. The method may include the steps of receiving input data at an input of the computational device, generating a temporal prediction based on the input data, generating a geospatial prediction based on the input data, and generating output data based on the time series and geospatial predictions. The output data may be configured to display at least one user interface at an output of the computational device. |
申请公布号 |
US8924332(B2) |
申请公布日期 |
2014.12.30 |
申请号 |
US201013322626 |
申请日期 |
2010.05.28 |
申请人 |
Purdue Research Foundation |
发明人 |
Maciejewski Ross;Hafen Ryan;Rudolph Stephen;Cleveland William;Ebert David |
分类号 |
G06F17/00;G06N5/02;G06Q10/10;G06F17/18 |
主分类号 |
G06F17/00 |
代理机构 |
Maginot, Moore & Beck, LLP |
代理人 |
Maginot, Moore & Beck, LLP |
主权项 |
1. A method for forecasting hotspots on a computational device, comprising the steps of:
receiving input data at an input of the computational device; generating a temporal prediction based on the input data, wherein the temporal prediction is based at least partially on a seasonal trend decomposition model based on locally weighted scatterplot smoothing (LOESS) applied to the equation √{square root over (Yt)}=Tt,+St+Dt+rt where for the t-th day Yt is the time series for the temporal prediction, Tt is the inter-annual component, St is the yearly-seasonal component, Dt is the day-of-the-week component and rt is the remainder; generating a geospatial prediction based on the input data; and generating output data based on the time series and the geospatial prediction, the output data being configured to display at least one user interface at an output of the computational device. |
地址 |
West Lafayette IN US |