Machine learning (ML) is a subject that focuses on the data analysis using various statistical to... more Machine learning (ML) is a subject that focuses on the data analysis using various statistical tools and learning processes in order to gain more knowledge from the data. The objective of this research was to apply one of the ML techniques on the low birth weight (LBW) data in Indonesia. This research conducts two ML tasks, including prediction and classification. The binary logistic regression model was firstly employed on the train and the test data. Then, the random approach was also applied to the data set. The results showed that the binary logistic regression had a good performance for prediction , but it was a poor approach for classification. On the other hand, random forest approach has a very good performance for both prediction and classification of the LBW data set.
The birth interval of first child is one example of survival analysis. The objective of this rese... more The birth interval of first child is one example of survival analysis. The objective of this research is to determine the best Cox regression model that describes the impact of several factors to the birth interval of first child in South Sumatra Province. The data were based on the Indonesia Demographic and Health Survey (IDHS) 2012. In this work, we included four factors, i.e. the age of the mother, mother's highest education, place of residence, and wealth index. Based on several tests that have been conducted, the factors which significantly affect the survival time were the age of the mother, mother's highest education (junior high school), and wealth index (middle and very rich).
Region and time effects specification of the panel data can explain the difference among the valu... more Region and time effects specification of the panel data can explain the difference among the values of Human Development Index (HDI) in each region and period of time. The objectives of this study are (1) to determine the best panel data regression model with one way component effect in explaining the diversity of HDI values, and (2) to analyze the factors that significantly affect the changes of the HDI values. The data used in this research are the HDI values of all districts and cities in South Sumatra Province from year 2007 to 2014 which obtained from the Statistics of Sumatera Selatan Province. By using the Chow test, Hausman test, and the selection of the one way component effects, the results show that the best panel data regression model for HDI values in South Sumatra Province from year 2007 to 2014 is time component fixed effect model with seemingly unrelated regression weighted. Meanwhile, the factors that significantly affect the changing of the HDI values in South Sumatra Province from year 2007 to 2014 are the number of households with clean water access, senior high school participation rate, the literacy rate, and the labor force participation rate.
The spread of tuberculosis (TB) among individuals in the population can be described by the epide... more The spread of tuberculosis (TB) among individuals in the population can be described by the epidemic SEIR model, which is a mathematical model that divides the population into four subpopulations i.e. susceptible (S), exposed (E), infected (I), and recovered (R). The objective of this research is to build an epidemic SEIR model for TB transmission by involving total therapy rate (R0) in infected subpopulation. To illustrate the effects of R0, a numerical simulation with different values of R0 was also carried out using R software. The results showed that the greater value of the total therapy rate, the decrease in the number of in- dividuals in infected subpopulation became faster.
In the prepaid electricity system, the users have to pay in advance for the purchase of electrica... more In the prepaid electricity system, the users have to pay in advance for the purchase of electrical energy in the form of a voucher code. The code should be well protected and difficult to be cracked by anyone. The main objective of this research is to combine DES algorithm and RSA Algorithm for enhancing the security system of electricity voucher code. The code in this article was based on individual information such as the electricity meter number, the number of pulses of electricity used, the date and time of purchase. Such individual information would generate the numbers that were encrypted using the DES algorithm. Furthermore, some characters of the chipertext from DES algorithm became the plaintext in RSA algorithm. As the result, the plaintext was encrypted again using RSA algorithm to gain a new electricity voucher code.
Tuberculosis is one of the infectious diseases that caused by Mycobacterium tuberculosis (Mtb). T... more Tuberculosis is one of the infectious diseases that caused by Mycobacterium tuberculosis (Mtb). The aim of this research is to analyze the stability of the Susceptible InfectedRecovered (SIR) model for tuberculosis transmission. We firstly constructed the epidemic SIR model and afterwards derived disease-free equilibrium, endemic equilibrium, and the basic reproduction number. Based on the analysis, both disease-free equilibrium and endemic equilibrium for the developed SIR model were stable.Finally, a numerical example wasalso given in support of the result.
In this study, we aimed to (1) show whether the Sriwijaya University tracer study data follow som... more In this study, we aimed to (1) show whether the Sriwijaya University tracer study data follow some survival distributions, (2) find the best survival distribution to represent the data, and (3) estimate the survival probability and hazard rate of the data. The tracer study was conducted from January 1, 2012 to December 31, 2012. There were 637 alumni who participated in the study. The result showed that the data follow the normal distribution, logistic distribution, and SEV distribution, in which the normal distribution was the best in representing the data. Based on the estimation procedure, the lowest probability of finding the first job was before graduation and the highest probability was about two years after graduation.
Time to get the first job data is one of important information for some University to make their ... more Time to get the first job data is one of important information for some University to make their academic policy. Usually, such information is obtained by a tracer study. That is a study about the most recent situation of the alumni’s job. When conducting a tracer study, the exact time to get the first job is often unknown. Instead, we just only know the interval in which the event has occurred. In survival analysis, this data is known as interval censored data. As a consequence, some conventional methods such as life table and Kaplan-Meier are no more appropriate to estimate survival functions. The main goal of this study was to apply some appropriate survival methods for interval censored data which obtained from a tracer study of Sriwijaya University in 2012. The instrument in collecting the data was questionnaires that can be accessed by the respondent on the website of Sriwijaya University. There were 637 alumni who participated as respondent in the tracer study. Respondent characteristics which included in the analysis were sex, faculty, graduation year, the information source about the job, consideration of taking the job, and taking additional course before getting the job. In order to estimate the survival probability, we use nonparametric maximum likelihood estimation. Based on the estimation, the highest probability of getting the first job was in the 24th month after the graduation day, meanwhile the lowest probability was before the graduation day. Then, we used Finkelstein’s test, Sun’s test, and Wilcoxon-Mann-Whitney test to compare the survival functions among categories. The result showed that the conclusions of the three tests do not have much difference.
Survival analysis is one of the topics in the field of mathematics which deals with statistical a... more Survival analysis is one of the topics in the field of mathematics which deals with statistical analysis of the time until the occurrence of one or more a particular event. The purposes of this study are to obtain the survival model and to estimate the parameters of the type I censored data which log-logistic distributed. Maximum likelihood estimate was used to estimate the unknown parameters. Based on the results and discussion, if the exact values of the parameters γ and β are analytically difficult to obtain, then the numerical approach by the Newton-Rhapson iteration can be used to approach the values of both parameters.
This study aims to (1) determine the probability of graduate’s time to get the first job, (2) kno... more This study aims to (1) determine the probability of graduate’s time to get the first job, (2) know how the influence of some characteristics toward graduate’s time to get to first job by applying Cox proportional hazard model. Research subjects consisted of 35 graduates of the Department of Mathematics, Faculty of Mathematics and Natural Sciences, Sriwijaya University who graduated in 2012. Observable characteristics are age, period of study, GPA, TOEFL scores, parental education, organizational experience, and work experience. The results of the study showed that the highest probability to get the first job is start at the beginning of the third month until the end of the fourth month after the graduation day, that is equal to 0,3159. The characteristic which significantly affects the time to get the first job is organizational experience. The best formed model is h(t,X)= h_0 (t) exp(-0,979 X_6) with a hazard ratio equal to 0,376. This means that the graduate who have organizational experiences has a chance 0,376 times more likely to get that first job after graduation.
This study aims to (1) produce a valid and practical interactive learning media based Lectora Ins... more This study aims to (1) produce a valid and practical interactive learning media based Lectora Inspire in a discrete method course, (2) see the effects of the developed learning media to the students’ learning outcome in a discrete method class. Research and development, which consists of five stages: predevelopment, development, evaluation, field test, and revision, was implemented in this study. The developed interactive learning media was tested against 23 students of discrete method class in the Department of Mathematics, Faculty of Mathematics and Natural Science, Sriwijaya University. The results of the analysis are: (1) based on the evaluation of the experts and the perceptions of the students, the interactive learning media which has been developed was valid, practical, and highly recommended to be used as a source of student’s self-learning in discrete method courses, (2) the interactive learning media which has been developed was effective to increase students’ learning outcome.
Survival analysis is a branch of statistics, which is focussed on the analysis of time-to-event d... more Survival analysis is a branch of statistics, which is focussed on the analysis of time-to-event data. In multivariate survival analysis, the proportional hazards (PH) is the most popular model in order to analyze the effects of several covariates on the survival time. However, the assumption of constant hazards in PH model is not always satisfied by the data. The violation of the PH assumption leads to the misinterpretation of the estimation results and decreasing the power of the related statistical tests. On the other hand, the accelerated failure time (AFT) models do not assume the constant hazards in the survival data as in PH model. The AFT models, moreover, can be used as the alternative to PH model if the constant hazards assumption is violated. The objective of this research was to compare the performance of PH model and the AFT models in analyzing the significant factors affecting the first birth interval (FBI) data in Indonesia. In this work, the discussion was limited to three AFT models which were based on Weibull, exponential, and log-normal distribution. The analysis by using graphical approach and a statistical test showed that the non-proportional hazards exist in the FBI data set. Based on the Akaike information criterion (AIC), the log-normal AFT model was the most appropriate model among the other considered models. Results of the best fitted model (log-normal AFT model) showed that the covariates such as women's educational level, husband's educational level, contraceptive knowledge, access to mass media, wealth index, and employment status were among factors affecting the FBI in Indonesia.
Machine learning (ML) is a subject that focuses on the data analysis using various statistical to... more Machine learning (ML) is a subject that focuses on the data analysis using various statistical tools and learning processes in order to gain more knowledge from the data. The objective of this research was to apply one of the ML techniques on the low birth weight (LBW) data in Indonesia. This research conducts two ML tasks, including prediction and classification. The binary logistic regression model was firstly employed on the train and the test data. Then, the random approach was also applied to the data set. The results showed that the binary logistic regression had a good performance for prediction , but it was a poor approach for classification. On the other hand, random forest approach has a very good performance for both prediction and classification of the LBW data set.
The birth interval of first child is one example of survival analysis. The objective of this rese... more The birth interval of first child is one example of survival analysis. The objective of this research is to determine the best Cox regression model that describes the impact of several factors to the birth interval of first child in South Sumatra Province. The data were based on the Indonesia Demographic and Health Survey (IDHS) 2012. In this work, we included four factors, i.e. the age of the mother, mother's highest education, place of residence, and wealth index. Based on several tests that have been conducted, the factors which significantly affect the survival time were the age of the mother, mother's highest education (junior high school), and wealth index (middle and very rich).
Region and time effects specification of the panel data can explain the difference among the valu... more Region and time effects specification of the panel data can explain the difference among the values of Human Development Index (HDI) in each region and period of time. The objectives of this study are (1) to determine the best panel data regression model with one way component effect in explaining the diversity of HDI values, and (2) to analyze the factors that significantly affect the changes of the HDI values. The data used in this research are the HDI values of all districts and cities in South Sumatra Province from year 2007 to 2014 which obtained from the Statistics of Sumatera Selatan Province. By using the Chow test, Hausman test, and the selection of the one way component effects, the results show that the best panel data regression model for HDI values in South Sumatra Province from year 2007 to 2014 is time component fixed effect model with seemingly unrelated regression weighted. Meanwhile, the factors that significantly affect the changing of the HDI values in South Sumatra Province from year 2007 to 2014 are the number of households with clean water access, senior high school participation rate, the literacy rate, and the labor force participation rate.
The spread of tuberculosis (TB) among individuals in the population can be described by the epide... more The spread of tuberculosis (TB) among individuals in the population can be described by the epidemic SEIR model, which is a mathematical model that divides the population into four subpopulations i.e. susceptible (S), exposed (E), infected (I), and recovered (R). The objective of this research is to build an epidemic SEIR model for TB transmission by involving total therapy rate (R0) in infected subpopulation. To illustrate the effects of R0, a numerical simulation with different values of R0 was also carried out using R software. The results showed that the greater value of the total therapy rate, the decrease in the number of in- dividuals in infected subpopulation became faster.
In the prepaid electricity system, the users have to pay in advance for the purchase of electrica... more In the prepaid electricity system, the users have to pay in advance for the purchase of electrical energy in the form of a voucher code. The code should be well protected and difficult to be cracked by anyone. The main objective of this research is to combine DES algorithm and RSA Algorithm for enhancing the security system of electricity voucher code. The code in this article was based on individual information such as the electricity meter number, the number of pulses of electricity used, the date and time of purchase. Such individual information would generate the numbers that were encrypted using the DES algorithm. Furthermore, some characters of the chipertext from DES algorithm became the plaintext in RSA algorithm. As the result, the plaintext was encrypted again using RSA algorithm to gain a new electricity voucher code.
Tuberculosis is one of the infectious diseases that caused by Mycobacterium tuberculosis (Mtb). T... more Tuberculosis is one of the infectious diseases that caused by Mycobacterium tuberculosis (Mtb). The aim of this research is to analyze the stability of the Susceptible InfectedRecovered (SIR) model for tuberculosis transmission. We firstly constructed the epidemic SIR model and afterwards derived disease-free equilibrium, endemic equilibrium, and the basic reproduction number. Based on the analysis, both disease-free equilibrium and endemic equilibrium for the developed SIR model were stable.Finally, a numerical example wasalso given in support of the result.
In this study, we aimed to (1) show whether the Sriwijaya University tracer study data follow som... more In this study, we aimed to (1) show whether the Sriwijaya University tracer study data follow some survival distributions, (2) find the best survival distribution to represent the data, and (3) estimate the survival probability and hazard rate of the data. The tracer study was conducted from January 1, 2012 to December 31, 2012. There were 637 alumni who participated in the study. The result showed that the data follow the normal distribution, logistic distribution, and SEV distribution, in which the normal distribution was the best in representing the data. Based on the estimation procedure, the lowest probability of finding the first job was before graduation and the highest probability was about two years after graduation.
Time to get the first job data is one of important information for some University to make their ... more Time to get the first job data is one of important information for some University to make their academic policy. Usually, such information is obtained by a tracer study. That is a study about the most recent situation of the alumni’s job. When conducting a tracer study, the exact time to get the first job is often unknown. Instead, we just only know the interval in which the event has occurred. In survival analysis, this data is known as interval censored data. As a consequence, some conventional methods such as life table and Kaplan-Meier are no more appropriate to estimate survival functions. The main goal of this study was to apply some appropriate survival methods for interval censored data which obtained from a tracer study of Sriwijaya University in 2012. The instrument in collecting the data was questionnaires that can be accessed by the respondent on the website of Sriwijaya University. There were 637 alumni who participated as respondent in the tracer study. Respondent characteristics which included in the analysis were sex, faculty, graduation year, the information source about the job, consideration of taking the job, and taking additional course before getting the job. In order to estimate the survival probability, we use nonparametric maximum likelihood estimation. Based on the estimation, the highest probability of getting the first job was in the 24th month after the graduation day, meanwhile the lowest probability was before the graduation day. Then, we used Finkelstein’s test, Sun’s test, and Wilcoxon-Mann-Whitney test to compare the survival functions among categories. The result showed that the conclusions of the three tests do not have much difference.
Survival analysis is one of the topics in the field of mathematics which deals with statistical a... more Survival analysis is one of the topics in the field of mathematics which deals with statistical analysis of the time until the occurrence of one or more a particular event. The purposes of this study are to obtain the survival model and to estimate the parameters of the type I censored data which log-logistic distributed. Maximum likelihood estimate was used to estimate the unknown parameters. Based on the results and discussion, if the exact values of the parameters γ and β are analytically difficult to obtain, then the numerical approach by the Newton-Rhapson iteration can be used to approach the values of both parameters.
This study aims to (1) determine the probability of graduate’s time to get the first job, (2) kno... more This study aims to (1) determine the probability of graduate’s time to get the first job, (2) know how the influence of some characteristics toward graduate’s time to get to first job by applying Cox proportional hazard model. Research subjects consisted of 35 graduates of the Department of Mathematics, Faculty of Mathematics and Natural Sciences, Sriwijaya University who graduated in 2012. Observable characteristics are age, period of study, GPA, TOEFL scores, parental education, organizational experience, and work experience. The results of the study showed that the highest probability to get the first job is start at the beginning of the third month until the end of the fourth month after the graduation day, that is equal to 0,3159. The characteristic which significantly affects the time to get the first job is organizational experience. The best formed model is h(t,X)= h_0 (t) exp(-0,979 X_6) with a hazard ratio equal to 0,376. This means that the graduate who have organizational experiences has a chance 0,376 times more likely to get that first job after graduation.
This study aims to (1) produce a valid and practical interactive learning media based Lectora Ins... more This study aims to (1) produce a valid and practical interactive learning media based Lectora Inspire in a discrete method course, (2) see the effects of the developed learning media to the students’ learning outcome in a discrete method class. Research and development, which consists of five stages: predevelopment, development, evaluation, field test, and revision, was implemented in this study. The developed interactive learning media was tested against 23 students of discrete method class in the Department of Mathematics, Faculty of Mathematics and Natural Science, Sriwijaya University. The results of the analysis are: (1) based on the evaluation of the experts and the perceptions of the students, the interactive learning media which has been developed was valid, practical, and highly recommended to be used as a source of student’s self-learning in discrete method courses, (2) the interactive learning media which has been developed was effective to increase students’ learning outcome.
Survival analysis is a branch of statistics, which is focussed on the analysis of time-to-event d... more Survival analysis is a branch of statistics, which is focussed on the analysis of time-to-event data. In multivariate survival analysis, the proportional hazards (PH) is the most popular model in order to analyze the effects of several covariates on the survival time. However, the assumption of constant hazards in PH model is not always satisfied by the data. The violation of the PH assumption leads to the misinterpretation of the estimation results and decreasing the power of the related statistical tests. On the other hand, the accelerated failure time (AFT) models do not assume the constant hazards in the survival data as in PH model. The AFT models, moreover, can be used as the alternative to PH model if the constant hazards assumption is violated. The objective of this research was to compare the performance of PH model and the AFT models in analyzing the significant factors affecting the first birth interval (FBI) data in Indonesia. In this work, the discussion was limited to three AFT models which were based on Weibull, exponential, and log-normal distribution. The analysis by using graphical approach and a statistical test showed that the non-proportional hazards exist in the FBI data set. Based on the Akaike information criterion (AIC), the log-normal AFT model was the most appropriate model among the other considered models. Results of the best fitted model (log-normal AFT model) showed that the covariates such as women's educational level, husband's educational level, contraceptive knowledge, access to mass media, wealth index, and employment status were among factors affecting the FBI in Indonesia.
Uploads
Papers by Alfensi Faruk
probability and hazard rate of the data. The tracer study was conducted from January 1, 2012 to December 31, 2012. There were 637 alumni who participated in the study. The result showed that the data follow the normal distribution, logistic distribution, and SEV distribution, in which the normal distribution was the best in representing the data. Based on the estimation procedure, the lowest probability of finding the first job was before graduation and the highest probability was about two years after graduation.
Conference Presentations by Alfensi Faruk
probability and hazard rate of the data. The tracer study was conducted from January 1, 2012 to December 31, 2012. There were 637 alumni who participated in the study. The result showed that the data follow the normal distribution, logistic distribution, and SEV distribution, in which the normal distribution was the best in representing the data. Based on the estimation procedure, the lowest probability of finding the first job was before graduation and the highest probability was about two years after graduation.