Comparison of Machine Learning Algorithms for Discharge Prediction of Multipurpose Dam

For effective water management in the downstream area of a dam, it is necessary to estimate the amount of discharge from the dam to quantify the flow downstream of the dam. In this study, a machine learning model was constructed to predict the amount of discharge from Soyang River Dam using precipitation and dam inflow/discharge data from 1980 to 2020. Decision tree, multilayer perceptron, random forest, gradient boosting, RNN-LSTM, and CNN-LSTM were used as algorithms. The RNN-LSTM model achieved a Nash–Sutcliffe efficiency (NSE) of 0.796, root-mean-squared error (RMSE) of 48.996 m3/s, mean absolute error (MAE) of 10.024 m3/s, R of 0.898, and R2 of 0.807, showing the best results in dam discharge prediction. The prediction of dam discharge using machine learning algorithms showed that it is possible to predict the amount of discharge, addressing limitations of physical models, such as the difficulty in applying human activity schedules and the need for various input data.

IMPROVEMENT OF MATHEMATICS LEARNING OUTCOMES ABOUT DISTANCE, TIME, AND SPEED THROUGH PROJECT BASED LEARNING (PjBL) MODEL TO FIFTH GRADE STUDENTS OF SD NEGERI KLEDUNGKRADENAN IN ACADEMIC YEAR OF 2020/2021

<p><strong><em>Abstract</em></strong></p><p><em>The objectives of the study were: (1) to describe the steps in implementing Project Based Learning (PjBL) model in improving mathematics learning outcomes about distance, time, and speed, (2) to increase mathematics learning outcomes about distance, time and speed by applying Project Based Learning (PjBL), and (3) to describe the constraints and solutions in implementing Project Based Learning (PjBL) model in improving mathematics learning outcomes about distance, time, and speed. The research was collaborative classroom action research conducted in three cycles. The subjects were students and teacher of fifth grade SDN Kledungkradenan. The data were qualitative and quantitative data. Data collection techniques used observation, interviews, and tests. The data validity used triangulation of techniques and triangulation of sources. Quantitative and qualitative data analysis included data reduction, presentation, and conclusion. The results showed that: (1) the steps in the implementation of Project Based Learning (PjBL) model were: (a) presenting problems, (b) preparing project plans, (c) preparing project activity schedules, (d) monitoring project implementation, (e) presenting project results, (f) evaluating project results; (2) the application of Project Based Learning (PjBL) model improved mathematics learning outcomes about distance, time and speed; (3) the constrains were: some students were passive in project activities and did not get enough tools needed in making projects while the solutions were: the teacher monitored the learning and discussion activities, the teacher encouraged the students to participate actively in project activities, and the teacher and the students prepared the tools and materials needed in learning</em></p>

A population data-driven workflow for COVID-19 modeling and learning

CityCOVID is a detailed agent-based model that represents the behaviors and social interactions of 2.7 million residents of Chicago as they move between and colocate in 1.2 million distinct places, including households, schools, workplaces, and hospitals, as determined by individual hourly activity schedules and dynamic behaviors such as isolating because of symptom onset. Disease progression dynamics incorporated within each agent track transitions between possible COVID-19 disease states, based on heterogeneous agent attributes, exposure through colocation, and effects of protective behaviors of individuals on viral transmissibility. Throughout the COVID-19 epidemic, CityCOVID model outputs have been provided to city, county, and state stakeholders in response to evolving decision-making priorities, while incorporating emerging information on SARS-CoV-2 epidemiology. Here we demonstrate our efforts in integrating our high-performance epidemiological simulation model with large-scale machine learning to develop a generalizable, flexible, and performant analytical platform for planning and crisis response.

Article Desktop Based National Police Commission Activities Information System

In the implementation of the activities of the National Police Commission (Kompolnas) every month it is necessary to prepare a monthly report by the Secretariat of the National Police Commission to be reported to the Chair of the National Police, so that it can be used as a benchmark for the implementation of Kompolnas performance. However, making monthly reports is often late and mistaken because there are factors that influence, including: data storage such as official travel administration and irregular activity schedules making it difficult to find data, difficult to monitor ongoing activities, many activities implemented and others. In this regard, a desktop-based information system for the implementation of Kompolnas is required. The research method uses Research and Development to create a desktop-based Kompolnas activity information system. Meanwhile, the data teaching technique used literature study and field study (interview and document examination). Then the system development method used is the waterfall method which has several stages, namely: needs analysis, system design, coding & testing, program implementation, and maintenance. The research resulted in a supporting information system for compiling a monthly report on Kompolnas activities because data on Kompolnas activities such as reference documents, orders, materials, photos / videos, are easy to input and can be viewed, downloaded and printed. In doing so, it can realize a paperless office, thus giving the printer paper, ink and time.

Pemberdayaan Ibu Rumah Tangga Melalui Pembuatan Handsanitizer Dalam Upaya Pencegahan Covid-19

Housewives have an important role to play in supporting government programs in preventing the spread of COVID-19 at the household level. Community service activities aim to provide understanding to partners so that they have knowledge and concern about preventing the spread of COVID-19 through the application of the 5M health protocol, increasing creativity in efforts to prevent the spread of COVID-19 through education on how to make hand sanitizers from natural ingredients of betel leaf (Piper betle Linn) which are mostly found in Sukorejo Village. The method used in this activity is counseling by providing material through lectures and distributing hand sanitizer finished products. The implementation of this service activity in the form of dissemination of the prevention of the spread of COVID-19 involves active participation from partners starting from the activity planning process, preparation of activity schedules to education on how to make and distribute hand sanitizers. This activity generated ideas for partners who are mostly housewives to empower themselves in supporting government programs to prevent the spread of COVID-19 by practicing how to make natural hand sanitizers from betel leaf. Thus, the expected output from this activity is an increase in IRT knowledge in preventing the spread of COVID-19, how to manufacture, packaging hand sanitizers, and participants to obtain sample products which can increase interest in producing similar products and commitment to implementing health protocols in preventing the spread of COVID-19.

Reducing levels of Violence in the Psychiatric Intensive Care Unit (PICU) - a multidisciplinary quality improvement project

AimsOur aim: To reduce the number of Level 1* violent incidents in Ward 4 by 30% by April 2020*Level 1 is defined as “Behaviour involving force, which causes or is intended to cause physical harm to others; but excludes assault on objects, threats or verbal abuse”Ward 4 is Belfast Health and Social Care Trust's only PICU, with a total of 6 beds. Our project took place on the background of a recent move to a new purpose-built inpatient unit, as well as a trust-wide initiative to address levels of violence across inpatient psychiatry services.MethodWe divided our project into 3 main areas:Patient factorsStaff factorsEnvironmental factorsWe identified and implemented a number of change ideas, using Plan-Do-Study- Act methodology, regularly meeting to review progress and plotting our data on a run chart.Key patient interventions included a “Mutual Respect” exercise and regular “Community Meetings”.Staff interventions included use of Safety Crosses, Daily Safety Briefings and the Broset Violence Checklist (BVC).Environmental factors were continually assessed and escalated as appropriate.We raised awareness of our project and gained feedback by creating a dedicated notice board, providing a staff information session and including it as an agenda item at ward meetings.Our project measures were identified as:Outcome: Number of level 1 violent incidents occurring per weekBalancing: Number of incidents in other categories; Patient satisfactionProcess: Staff safety rating; Engagement with interventionsResultUnfortunately, we were unable to meet our initial goal and there continued to be considerable variation in the number of weekly incidents.We believe this was attributable to several factors, including the level of acuity within the ward during the project timeframe. It was noted that a relatively small number of patients contributed to a large proportion of the total incidents. Our results, therefore, did not reflect the success of interventions with other patients on the ward.Despite this, we noted improvements in terms of patient and staff engagement with the project, including subjective reports of staff safety during shifts.ConclusionThe unpredictable and complex nature of the PICU setting cannot be under-estimated and this ultimately impacted on achieving our intended outcome.We do feel, however, that the project has had a positive impact and we hope we can build on this progress over the coming months.Further interventions are being explored, including personalised daily activity schedules and attempts to reduce levels of continuous observations.

Behavior skills training for teaching and generalizing self-instruction skills for students with intellectual disability

Research demonstrates that combining video modeling and visual activity schedules into video activity schedules has been effective in increasing independence for students with disabilities. However, the instructional procedures used to teach students to acquire the necessary skills to navigate those tools vary among existent research. In this study, a behavior skills training package was investigated within a multiple probe design across participants, to determine if four elementary aged participants with intellectual disability could acquire self-instruction skills to navigate a video activity schedule. Three participants acquired the self-instruction skills using behavior skills training. After learning how to navigate the video activity schedules, the same three participants generalized and maintained the self-instruction skills to perform additional novel behaviors. The results suggest that behavior skills training may be an effective instructional strategy for teaching self-instruction skills and a means to program for generalized behavior change for students with intellectual disability.

THE EFFECT OF ACTIVITY SCHEDULES TO TEACH APPROPRIATE ACTIVITIES DURING FREE TIME TO STUDENTS WITH AUTISM SPECTRUM DISORDERS

Background: Recent literature suggests that activity schedules increase engagement and independent play skills while decreasing interfering behaviors for students with autism. Therefore, in this study, four students with diagnosed with autism spectrum disorder were taught to use an activity schedules at free time to examine the effects of it on rate of self-stimulatory behavior. Methods: A multiple baseline design was used to examine these effects. The three students were given an activity schedule housed in a binder with five different recess activities. Activities remained constant throughout the study, but the order was changed. Students followed the activity schedule while experimenters tracked frequency of self-stimulatory behaviors. The frequency was then divided by the duration it took the student to complete the schedule to produce rate data. Findings: Results showed that all three students had high levels of self-stimulatory behavior prior to implementation of the schedule (range of subjects: 6.3 – 10.5 times per minute). With the implementation of the activity schedule, all three students had significantly decreased rates of self-stimulatory behavior (range of subjects: (1.5 – 2.2 times per minute). The consistent results show that activity schedules decrease the rate of self-stimulatory behavior at recess. Conclusion: These findings support previous research performed over the last several years and confirmed the effectiveness of activity schedules for students with autism. <p> </p><p><strong> Article visualizations:</strong></p><p><img src="/-counters-/edu_01/0747/a.php" alt="Hit counter" /></p>