Application of Hierarchical Clustering Algorithm to Evaluate Students Performance of an Institute

Rana, Shiwani; Garg, Roopali

doi:10.1109/cict.2016.143

Cited by 19 publications

(9 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the use of clustering methods, we built on earlier studies [7,22]. Though there exist many clustering methods, we used the MCM, which has also been extensively used in the literature [23,24]. The main advantages are that it relies on statistical models and requires no pre-specified number of clusters [25].…”

Section: Discussionmentioning

confidence: 99%

Introducing an efficient sampling method for national surveys with limited sample sizes: application to a national study to determine quality and cost of healthcare

et al. 2021

View full text Add to dashboard Cite

Background Sampling a small number of participants from an entire country is not straightforward. In this case, researchers reluctantly sample from a single setting or few settings, which limits the generalizability of findings. Therefore, there is a need to design efficient sampling method for small sample size surveys that can produce generalizable results at the country level. Methods Data comprised of twenty proxy variables to measure health services demands, structures, and outcomes of 413 districts of Iran. We used two data mining methods (hierarchical clustering method (HCM) and model-based clustering method (MCM)) to create homogenous groups of districts, i.e., strata based on these variables. We compared the internal and stability validity of the methods by statistical indices. An expert group checked the face validity of the methods, particularly regarding the total number of strata and the combination of districts in each stratum. The efficiency of selected method, which is measured by the inverse of variance, was compared with a simple random sampling (SRS) through simulation. The sampling design was tested in a national study in Iran, which aimed to evaluate the quality and costs of medical care for eight selected diseases by only recruiting 300 participants per disease at the country level. Results MCM and HCM divided the districts into eight and two clusters, respectively. The measures of internal and stability validity showed that clusters created by MCM were more separated, compact, and stable, thus forming our optimum strata. The probability of death from stroke, chronic obstructive pulmonary disease, and in-hospital mortality rate were the most important indicators that distinguished the eight strata. Based on the simulation results, MCM increased the efficiency of the sampling design up to 1.7 times compared to SRS. Conclusions The use of data mining improved the efficiency of sampling up to 1.7 times greater than SRS and markedly reduced the number of strata to eight in the entire country. The proposed sampling design also identified key variables that could be used to classify districts in Iran for sampling from these target populations in the future studies.

show abstract

Section: Discussionmentioning

confidence: 99%

Introducing an efficient sampling method for national surveys with limited sample sizes: application to a national study to determine quality and cost of healthcare

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Challenges faced by each method are generally stated in the advantages and disadvantages previous research. Predictive results using classification and cluster methods by previous researchers only predict students according to the predetermined class, not following the performance of the individuals involved [34], [35]. Many researchers could not distinguish between classification and cluster methods.…”

Section: Predictive Models Used In Previous Studies and Resultsmentioning

confidence: 99%

Systematic Research on Predictive Models on Students’ Academic Performance in Higher Education

2019

ijrte

View full text Add to dashboard Cite

In Malaysia, the students’ academic assessment has changed over time as it is now made to be more holistic. Predictive models used for educational data in Malaysia are still inadequate in giving a true picture of the students’ academic performance. The reason behind this is due to the lack of research in Malaysian context on predictive models and factors affecting students' academic achievement. Therefore, this study will conduct a systematic review on the literature, in order to identify the predictive methods for students’ academic performance in higher education. Through three major bases: WoS, Science Direct and IEEE Xplore, an extensive search was conducted to find any related articles from 2014 to 2018 that use "predict", "forecast", "academic performance", "student" and "higher education" keywords in their text. Initially, 195 articles were selected to be used in this review. However, through titles and abstracts’ filtering and screen processing, only 69 articles found to discuss on the predictive model for student's academic performance at higher education level. Based on extensive reading, the most widely used attributes in the predictive models is the academic process. While predictive models can be categorized into three, namely classification, cluster and regression, there are nine methods used by previous researchers in all categories. The most widely used category from previous studies was the classification, with 33 articles. This study has listed the advantages and disadvantages of each method based on the previous studies. This study also has identified the challenges and gaps faced by previous researchers in improving the existing models in the future. Among the challenges faced by the predictive models are the amount of data and assumptions that need to be followed before analysis can be made. In the future, the predictive models used for students’ academic performance should consider the latest assessments’ valuation method based on a modern educational system that emphasizes on soft skills, interpersonal skills and high-level thinking capabilities.

show abstract

“…It relies on statistical models and requires no pre-specified number of clusters [24]. HCM has also been extensively used in the literature [25,26].…”

Section: Discussionmentioning

confidence: 99%

Introducing an Efficient Sampling Method for National Surveys with Limited Sample Sizes: Application to a National Study to Determine Quality and Cost of Healthcare.

Parsaeian¹,

Mahdavi

Saadati

et al. 2020

Preprint

View full text Add to dashboard Cite

Background: Due to budget constraints, many national studies are only affordable with small sample sizes. Facing a similar constraint, the national study of “Iran Quality of Care in Medicine Program” (IQCAMP), aims to evaluate the quality and costs of medical care in the country for 8 selected high-cost high-volume diseases by only recruiting 300 participants per disease under study. We proposed and tested a data mining method to efficiently design such national studies with small sample sizes.Methods: We developed a sampling design based on hierarchical clustering method (HCM) and model-based clustering method (MCM). Based on these methods, we used healthcare structure and outcome data to define the optimum clusters of districts and provinces of Iran. An expert group checked the face validity of the defined clusters. We subsequently compared the internal validity of HCM with MCM by within clusters sum of square, average silhouette width, and Dunn index. The stability validity was also examined by statistical indices. We selected the optimum clusters based on the results of these measures. Per cluster, one province was systematically selected for data collection. The features of selected clusters were delineated by the Decision Tree Learning and the most important distinguishing indicators were identified. The efficiency of the selected sampling method was compared with a simple random sampling through simulation.Results: MCM and HCM divided the districts into eight and two clusters, respectively. The measures of internal and stability validity showed that clusters created by MCM were more separated, compact, and stable, thus forming our optimum clusters. The probability of death from stroke, chronic obstructive pulmonary disease, and in-hospital mortality rate were the most important indicators that distinguished the eight clusters. Based on the simulation results, MCM increased the efficiency of the sampling design up to 70% compared to simple random sampling.Conclusions: By using this sampling method we selected 300 participants per health condition across the country only from eight provinces rather than all 31 provinces. Our sampling method benefits studies with small sample sizes by decreasing the level of sampling variability and the overall cost of studies.

show abstract

Application of Hierarchical Clustering Algorithm to Evaluate Students Performance of an Institute

Cited by 19 publications

References 3 publications

Introducing an efficient sampling method for national surveys with limited sample sizes: application to a national study to determine quality and cost of healthcare

Introducing an efficient sampling method for national surveys with limited sample sizes: application to a national study to determine quality and cost of healthcare

Systematic Research on Predictive Models on Students’ Academic Performance in Higher Education

Introducing an Efficient Sampling Method for National Surveys with Limited Sample Sizes: Application to a National Study to Determine Quality and Cost of Healthcare.

Contact Info

Product

Resources

About