Forecasting Seizure Likelihood With Wearable Technology

The unpredictability of epileptic seizures exposes people with epilepsy to potential physical harm, restricts day-to-day activities, and impacts mental well-being. Accurate seizure forecasters would reduce the uncertainty associated with seizures but need to be feasible and accessible in the long-term. Wearable devices are perfect candidates to develop non-invasive, accessible forecasts but are yet to be investigated in long-term studies. We hypothesized that machine learning models could utilize heart rate as a biomarker for well-established cycles of seizures and epileptic activity, in addition to other wearable signals, to forecast high and low risk seizure periods. This feasibility study tracked participants' (n = 11) heart rates, sleep, and step counts using wearable smartwatches and seizure occurrence using smartphone seizure diaries for at least 6 months (mean = 14.6 months, SD = 3.8 months). Eligible participants had a diagnosis of refractory epilepsy and reported at least 20 seizures (mean = 135, SD = 123) during the recording period. An ensembled machine learning and neural network model estimated seizure risk either daily or hourly, with retraining occurring on a weekly basis as additional data was collected. Performance was evaluated retrospectively against a rate-matched random forecast using the area under the receiver operating curve. A pseudo-prospective evaluation was also conducted on a held-out dataset. Of the 11 participants, seizures were predicted above chance in all (100%) participants using an hourly forecast and in ten (91%) participants using a daily forecast. The average time spent in high risk (prediction time) before a seizure occurred was 37 min in the hourly forecast and 3 days in the daily forecast. Cyclic features added the most predictive value to the forecasts, particularly circadian and multiday heart rate cycles. Wearable devices can be used to produce patient-specific seizure forecasts, particularly when biomarkers of seizure and epileptic activity cycles are utilized.

Forecasting seizure likelihood with wearable technology

The unpredictability of epileptic seizures exposes people with epilepsy to potential physical harm, restricts day-to-day activities, and impacts mental well-being. Accurate seizure forecasters would reduce the uncertainty associated with seizures but need to be feasible and accessible in the long-term. Wearable devices are perfect candidates to develop non-invasive, accessible forecasts but are yet to be investigated in long-term studies. We hypothesized that machine learning models could utilize heart rate as a biomarker for well-established cycles of seizures and epileptic activity, in addition to other wearable signals, to forecast high and low risk seizure periods. This feasibility study tracked participants' (n = 11) heart rates, sleep, and step counts using wearable smartwatches and seizure occurrence using mobile seizure diaries for at least 6 months (mean = 14.6 months, SD = 3.8 months). Eligible participants had a diagnosis of refractory epilepsy and reported at least 20 seizures (mean = 135, SD = 123) during the recording period. An ensembled machine learning and neural network model estimated seizure risk either daily or hourly, with retraining occurring on a weekly basis as additional data was collected. Performance was evaluated retrospectively against a rate-matched random forecast using the area under the receiver operating curve. A pseudo-prospective evaluation was also conducted on a held-out dataset. Of the 11 participants, seizures were predicted above chance in all (100%) participants using an hourly forecast and in ten (91%) participants using a daily forecast. The average time spent in high risk (prediction time) before a seizure occurred was 37 minutes in the hourly forecast and 3 days in the daily forecast. Cyclic features added the most predictive value to the forecasts, particularly circadian and multiday heart rate cycles. Wearable devices can be used to produce patient-specific seizure forecasts, particularly when biomarkers of seizure and epileptic activity cycles are utilized.

Analysis of Employee Attrition using for Machine Learning Techniques

Nowadays, the daily forecast for employee losses becomes a major issue. Staff participation is an important issue for the organization, especially when professional technical staff and key people in the organization come from good positions. This leads to a loss of finances to replace skilled labor. Therefore, we use data from current and former employees to analyze common causes of employee access or influence. To avoid employment, we use several planning methods, namely: Decision Tree, Log-log of Backlog, SVM, KNN, Random Forest, Bayes Naive. To do this, we use the method to select employment information and analyze the results to avoid employee income. Companies need to anticipate employee incentives and contribute to economic growth by reducing manpower.

Seasonal and diurnal performance of daily forecasts with WRF V3.8.1 over the United Arab Emirates

Effective numerical weather forecasting is vital in arid regions like the United Arab Emirates (UAE) where extreme events like heat waves, flash floods, and dust storms are severe. Hence, accurate forecasting of quantities like surface temperatures and humidity is very important. To date, there have been few seasonal-to-annual scale verification studies with WRF at high spatial and temporal resolution. This study employs a convection-permitting scale (2.7 km grid scale) simulation with WRF with Noah-MP, in daily forecast mode, from 1 January to 30 November 2015. WRF was verified using measurements of 2 m air temperature (T2 m), 2 m dew point (TD2 m), and 10 m wind speed (UV10 m) from 48 UAE WMO-compliant surface weather stations. Analysis was made of seasonal and diurnal performance within the desert, marine, and mountain regions of the UAE. Results show that WRF represents temperature (T2 m) quite adequately during the daytime with biases ≤+1 ∘C. There is, however, a nocturnal cold bias (−1 to −4 ∘C), which increases during hotter months in the desert and mountain regions. The marine region has the smallest T2 m biases (≤-0.75 ∘C). WRF performs well regarding TD2 m, with mean biases mostly ≤ 1 ∘C. TD2 m over the marine region is overestimated, though (0.75–1 ∘C), and nocturnal mountain TD2 m is underestimated (∼-2 ∘C). UV10 m performance on land still needs improvement, and biases can occasionally be large (1–2 m s−1). This performance tends to worsen during the hot months, particularly inland with peak biases reaching ∼ 3 m s−1. UV10 m is better simulated in the marine region (bias ≤ 1 m s−1). There is an apparent relationship between T2 m bias and UV10 m bias, which may indicate issues in simulation of the daytime sea breeze. TD2 m biases tend to be more independent. Studies such as these are vital for accurate assessment of WRF nowcasting performance and to identify model deficiencies. By combining sensitivity tests, process, and observational studies with seasonal verification, we can further improve forecasting systems for the UAE.

Multi-step ahead time-series wind speed forecasting for smart-grid application

In this paper, an intelligent approach for short-term wind speed forecasting (STWSF) is proposed. The STWSF models are developed to forecast the wind speed into a multi-step ahead forecasting, which is used to demonstrate the daily forecast results in One-Step-Ahead (OSA), Two-Step-Ahead (TSA), and Three-Step-Ahead (ThSA) based forecasting manner. To demonstrate the performance and results of the proposed approach, the real-site logged dataset is used for training and testing phase of the year 2015 to 2017. The STWSF is achieved recursively by utilizing the forecasted data in step-1 (OSA) as an input to generate the next forecasting data (in step-2 TSA) and the process is achieved upto level of step-3 (ThSA) forecasting. In order to results demonstration of fair adoptability of the proposed approach, different neural networks (NNs) models are developed for the same dataset, which shows that the proposed STWSF approach is outperformed and can be utilized for other locations for future applications.

Peramalan Tren Penjualan Menu Restoran Menggunakan Metode Single Moving Average

<p>Penelitian ini bertujuan untuk meramalkan tren penjualan menu pada restoran guna membantu pihak pengelola restoran dalam menentukan dan memberikan rekomendasi pengelolaan stok menu. Peramalan dilakukan dengan mengimplementasikan metode <em>single moving average</em> pada data transaksi penjualan selama periode 15 bulan, yakni bulan Januari-Desember 2018 dan Januari-Maret 2019 untuk menghasilkan ramalan bulanan dan harian. Total sampel data latih yang diolah sebanyak 10.515 record yang merupakan data transaksi penjualan pada bulan Januari-Desember tahun 2018, serta 2.246 record data bulan Januari-Maret 2019 sebagai data uji (untuk menguji akurasi ramalan). Hasil pengujian hasil ramalan bulanan untuk Top-10 menu menghasilkan perhitungan MAPE <em>(Mean Absolut Percentage Error) sebesar </em>4% yang berarti tingkat akurasi sangat baik, yakni sebesar 96%. Sedangkan pengujian hasil ramalan harian menghasilkan MAPE yang cukup tinggi yaitu sebesar 39.2%, mengindikasikan nilai akurasi yang cukup rendah, yakni 60.8%. Meskipun akurasi untuk ramalan harian, masih rendah namun hasil penelitian ini dapat memberikan gambaran kepada pengelola hotel tentang rentang minimum-maksimal stok yang perlu disiapkan untuk menu tertentu pada hari-hari tertentu. Untuk memperoleh akurasi prediksi harian yang lebih akurat, penelitian ini akan dilanjutkan dengan mencoba metode lain serta menambah jumlah data latih.</p><p> </p><p class="Judul2"><strong><em>Abstract</em></strong></p><p class="Judul2"><em>This research aims to forecast sales trend of a restaurant menus to help the restaurant management in determaining and providing recommendations for managing stocks. Forecasting was performed by applying the single moving average towards fifteen months recorded data transaction, namely January to December 2018, and Januari to March 2019 to establish monthly and daily forecast. Total data training was 10.515 recods data transaction obtained from Januari to December 2018, while data testing was 2.246 record data transaction within Januari to March 2019. Result for montly forecast shows, that the average accuracy reached 96% (MAPE 4%) indicating the forecast is almost perfect. While, for daily forecast the average accuracy is only 60.8% (MAPE 39,2%) indicating that the forecast is less accurate. Although, accuracy of the daily forecast is considered less accurate, the result still can be used by the restaurant management to figure-out minimum and maximum amount of stock to be prepared for certain menus in certain days. </em></p>

Seasonal and diurnal performance of daily forecasts with WRF-NOAHMP V3.8.1 over the United Arab Emirates

Effective numerical weather forecasting is vital in arid regions like the United Arab Emirates (UAE) where extreme events like heat waves, flash floods and dust storms are severe. Hence, accurate forecasting of quantities like surface temperatures and humidity is very important. To date, there have been few seasonal-to-annual scale verification studies with WRF at high spatial and temporal resolution. This study employs a convection-permitting scale (2.7 km grid scale) simulation with WRF-NOAHMP, in daily forecast mode, from January 01 to November 30 2015. WRF was verified using measurements of 2 m air temperature (T-2m), dew point (TD-2m), and 10 m windspeed (UV-10m) from 48 UAE surface stations. Analysis was made of seasonal and diurnal performance within the desert, marine and mountain regions of the UAE. Results show that WRF represents temperature (T-2m) quite adequately during the daytime with biases ≤ +1 ˚C. There is however a nocturnal cold bias (−1 to −4 ˚C), which increases during hotter months in the desert and mountain regions. The marine region has the lowest T-2m biases (≤−0.75 ˚C). WRF performs well regarding TD-2m, with mean biases mostly ≤ 1 ˚C. TD-2m over the marine region is overestimated though (0.75–1 ˚C), and nocturnal mountain TD-2m is underestimated (~ −2 ˚C). UV-10m performance on land still needs improvement, and biases can occasionally be large (1–2 m s−1). This performance tends to worsen during the hot months, particularly inland with peak biases reaching ~ 3 m s−1. UV-10m are better simulated in the marine region (bias ≤ 1 m s−1). There is an apparent relationship between T-2m bias and UV-10m bias, which may indicate issues in simulation of the daytime sea breeze. TD-2m biases tend to be more independent. Studies such as these are vital for accurate assessment of WRF nowcasting performance and to identify model deficiencies. By combining sensitivity tests, process and observational studies with seasonal verification, we can further improve forecasting systems for the UAE.

Non-reasonable but efficient use of schemes in current model to improve realistic explicit convection modelling

&lt;p&gt;There is actually no limitation of current high-resolution weather model for producing simulation and forecast of convection at kilometer and infra-kilometer horizontal resolutions. However, the disappointing results as well as the associated huge amount of computer resources required may lead to focus on Large Eddy Simulation model instead. However, the use of LES is not trivial and required a long and non-portable adjustment over the region of interest. Also, it is difficult to use in operational mode for daily forecast since they require specific inputs.&lt;/p&gt;&lt;p&gt;In the other side, pushing the current regional or Limited Area Model towards very high resolution is a convenient way to reach explicit resolution of convective process for instance. However, an explicit simulation is not a guarantee of a realistic result mainly due to the fact that initial condition is crucial as well as all other descriptions of the environment (soil, vegetation, sst, etc) and use of correct parameterization schemes.&lt;/p&gt;&lt;p&gt;For instance, within the WRF model framework, one can identify more than 4000 set of parameterizations plus all the scheme adjustments and threshold associated to.&lt;/p&gt;&lt;p&gt;However, a physically based analyze of what it is necessary for a realistic and explicit convection simulation may conduct a physicist user to define its &amp;#8220;ideal&amp;#8221; physics with what it already exists in the model. It may conduct to so-called unrealistic model requests in term of computation requirement regarding the radiative, the turbulence and the microphysics schemes but it does works with HPC systems. This kind of parameterization will be presented here and used with a very realistic vertical circulation into convective systems with convective updraft and downdraft modelling, from few meters up to several kilometers height.&lt;/p&gt;

A near-term iterative forecasting system successfully predicts reservoir hydrodynamics and partitions uncertainty in real time

Freshwater ecosystems are experiencing greater variability due to human activities, necessitating new tools to anticipate future water quality. In response, we developed and deployed a real-time iterative water temperature forecasting system (FLARE – Forecasting Lake And Reservoir Ecosystems). FLARE is composed of: water quality and meteorology sensors that wirelessly stream data, a data assimilation algorithm that uses sensor observations to update predictions from a hydrodynamic model and calibrate model parameters, and an ensemble-based forecasting algorithm to generate forecasts that include uncertainty. Importantly, FLARE quantifies the contribution of different sources of uncertainty (driver data, initial conditions, model process, and parameters) to each daily forecast of water temperature at multiple depths. We applied FLARE to Falling Creek Reservoir (Vinton, Virginia, USA), a drinking water supply, during a 475-day period encompassing stratified and mixed thermal conditions. Aggregated across this period, root mean squared error (RMSE) of daily forecasted water temperatures was 1.13 C at the reservoir’s near-surface (1.0 m) for 7-day ahead forecasts and 1.62C for 16-day ahead forecasts. The RMSE of forecasted water temperatures at the near-sediments (8.0 m) was 0.87C for 7-day forecasts and 1.20C for 16-day forecasts. FLARE successfully predicted the onset of fall turnover 4-14 days in advance in two sequential years. Uncertainty partitioning identified meteorology driver data as the dominant source of uncertainty in forecasts for most depths and thermal conditions, except for the near-sediments in summer, when model process uncertainty dominated. Overall, FLARE provides an open-source system for lake and reservoir water quality forecasting to improve real-time management.Key PointsWe created a real-time iterative lake water temperature forecasting system that uses sensors, data assimilation, and hydrodynamic modelingOur water quality forecasting system quantifies uncertainty in each daily forecast and is open-source16-day future forecasted temperatures were within 1.4°C of observations over 16 months in a reservoir case study

METHODS OF SHORT-TERM FORECASTING OF ICE THICKNESS GROWTH IN THE NORTH-EASTERN PART OF THE CASPIAN SEA

The Caspian Sea is a southern sea with annual ice cover in the northern part. The thickness of the ice can reach one meter or more, depending on the severity of the winter. The sea ice of the Caspian Sea is characterized by significant variability, which affects human activities (industrial, fishing ones) as well as the fauna of the region. Based on daily information of North Caspian stations for the last 10 years, there has been developed short-term forecasting methodology for predicting daily increase in ice thickness. The effectiveness of the method was evaluated on the basis of calculation-dependent and independent materials of different lead times. The daily forecast of ice thickness growth was 82 to 98% justified. Climate research allowed us to establish that the maximum ice thickness has decreased stepwise since the late 1980s by 20–25 cm. This is due to the lower sum of negative temperatures, which in turn is associated with an increase in the number of days with a W form of atmospheric circulation and a decrease in the number of days with an E form in the winter period.