year. Prediction of Teacher Enrollment for Pakistani Schools by Using

 Abstract —Civilization distinguishes human being from other creatures. Civilization is adorned with education. Education uplifts not only the standard but also the historical records of nations. Teachers, students, public, government and curriculum are the main components of education. This study analyzes the number of school teachers and predicts the future annual enrollment of teachers using the SVM model (support vector machine). Time series data of 46 years (1971-2017) have been taken from the Handbook of Statistics on Pakistan Economy. Our results depict that school teachers must be further enrolled at all levels. Our method leads to good


I. INTRODUCTION
Human being is a masterpiece of creature. As the needs and demands of human being began to increase more and more, it started to discover the solutions of its problems. The human population went on growing more rapidly then small civilizations came into being. These civilizations were adorned with different gifts. The man acquired many qualities and values which categorized him into different standards of civilization. The main gift of nature which was bestowed upon him was to know what was right and what was wrong. What were the things which are beneficial to him? Who were his foes and friends? To answer all these questions a boon and blessing of nature knocked at his door. This knocker was a great gift in the form of education. No doubt education was a beauty for him. It was more than a golden ornament for him. It directed him towards the right path in the darkest clouds of ignorance and wilderness. Education grew with the passage of time for the betterment of humanity. Education developed and conquered civilization of the world. Manuscript  The study of Athens and Sparta which were old civilizations of the Greeks revealed that the secret of their beautification was the education. Undoubtedly teachers are super and supreme who guided the human beings. The focus point and the pivot around which the whole system of education revolved was the child. As a coordinator between the teacher and the tot (student) was the educational system but the logistics which were necessary for all the three afore-mentioned pillars were curricula. All these four pillars were dependable and correlator to each other. None of them could be denounced. All of them carry and bear history which was predated to be updated. In Pakistan education is divided into three tiers and all these tiers depend upon teachers. In order to standardize education, the quality of teachers is to be enhanced but in Pakistan quantity of the teachers is also enumerated along with quality. The analysis of 2005 and 2006 has shown that there was a great dropout percentage of the students due to many reasons and discrepancies. They may be on behalf of text books, teachers and department of education [1]. Pakistan has two types of education system, public and private, public institutions are under Pakistan's government. Public institutions in Pakistan cannot meet the quality of education due to the rapidly growing population with growth rate of 2.7% [2]. Hence, the government cannot fulfill the responsibility of providing education to the rapidly growing population [3]. In this study a statistical linear regression analysis and SVM model is applied in the education system to predict future enrollment of teachers in Pakistan. This method can be applied to all third world countries where educational institutions are less than their growing population. The third world countries desperately need new policies to increase enrollment of teachers for large population like Pakistan.
This research mainly focuses on predicting teacher enrollment at all levels, we use linear regression model and SVM to predict future teacher enrollment in Pakistan, time series data is analyzed over the time period of 46 years. As the data size is small so we use linear regression model and SVM accuracy having accuracy rate 61% and 73% respectively. The dataset contains three types of schools containing separate schools for boys, girls and coeducational schools. As there are separate male and female schools in Pakistan and both types of schools are less than the population so the growing population needs more male and female schools. A comprehensive sketch of the state of Pakistani schools enrolled teachers growing at all levels (primary schools, middle schools, high schools) is provided to contextualize the results and endorsements presented in this article and predicts how school teachers enrolled are increasing each year.  [10]. This work introduces an automatic framework to eliminate artificial participation and improve the prediction accuracy. Fig. 1 shows the overall flowchart of the proposed initialization method. The process consists of two major steps. The first step is to use static methods to divide the pre-process data into several stages. As the SVM algorithm allows annually enrolled teachers to have predictive relationships with multiple schools with varying degrees of membership based methods, therefore it is more reasonable in applications for educational institutions. Fig. 1 shows the yearly flow of teacher enrollment.   (0), middle school output (0), and secondary school output (0). Detailed description of the data is given in Table Ⅰ.

III. RESULTS
In this section, inclusive experiment is presented using data set designed based on teachers enrolled per year, acquired from Pakistan Economy Statistics Handbook as described in data collection. The dataset consists of the record of enrolled teachers in Pakistan over 46 years . Therefore we have a feature matrix of 46 rows over 3 columns and the accuracy is 61 percent and 73 percent. Fig. 2 shows the predicted data and original data, two methods of LR and SVM (support vector machine) are used in the predicted data. Primary education is seen as one of the main vehicles for promoting economic growth and improving living standards in developing countries [11], but it is likely to improve the quality of education if primary education involves new knowledge and adequate technical training at the primary level. Expanding access to primary education is a widely accepted priority in the fight against poverty. It is the first stage of compulsory education that establishes the academic foundation of the students and is considered as a fundamental right of all people. Primary education is highly linked to public [12] and private [13]- [15] institutions and the Express Tribune [16]- [19]. In this context,  Fig. 2 shows the teachers enrolled in middle schools per year and their future prediction. Fig. 2. This figure shows the three types of schools in Pakistan and comparison of two methods LR and SVM, and the prediction accuracy of SVM is higher than LR. Fig. 2(a) shows the primary schools and their prediction LR and SVM, Fig. 2(b) shows the middle schools and their prediction LR and SVM, Fig. 2(c) shows the high schools and their prediction LR and SVM. Fig. 3. This figure shows the three types of schools in Pakistan and SVM predicted data. Fig. 3 (a) shows primary school predicted data and Fig. 3 (b) shows middle school predicted data and Fig. 3(c) shows high school predicted data.
With annual examinations, high schools cover grades ninth to tenth. The students receive a secondary school diploma upon completion of grade 10. If they have the desire to continue inter-college further education, they must subsequently pass the high school examination.   Fig. 2 shows the total number of teachers enrolled in primary, middle and secondary schools from 1971 to 2017 and the prediction of enrolled teachers in primary, middle and high schools from 1971 to 2035. The teachers enrolled in high schools and their future predictions are shown in Fig. 2 per year. In Fig. 3, the SVM (Support Vector Method) and Table Ⅱ predicted data show the accuracy of both LR (Linear Regression) and SVM methods and the accuracy of both methods is 61% and 73% respectively. It proves that SVM presents the better prediction on this data set. We used two different methods LR and SVM for prediction and their respective accuracy are shown in Table  Ⅱ.

IV. CONCLUSION
The stakeholders of education are the teachers, students, educational system and the curricula. The development of a country mostly depends of its educational standards. In order to establish standard in education we need competent, efficient, and proficient and devoted teachers. The teachers are considered to be guides, motivators, revolutioners, reformers, the resource persons, facilitators, apostles, messengers, prophets, directors, educators, game changers, fate changers and character builders. In order to develop education for the children, we need female teachers due to having motherly nature. It is an admitted fact that women can perform vital role in the development of education of kids. The number of teachers in Pakistan is lower than the growing population. The predicted results show that the number of registered teachers is less than the population rate in 2035. Pakistan needs to develop new teacher enrolment policies and provide free basic education for every child. The results recommend that more female teachers should be enrolled in schools at all levels in the future, as in Pakistan the female population is 52 percent of the total population, while the female ratio is less than male teachers. Depending on the female population, additional female teachers are required. Recent progress in technology has altered the learning behaviors of students; besides giving a new impulse it reshapes the education itself. It can be said easily that the improvements in technologies empower students to learn more efficiently, effectively and contentedly.

CONFLICT OF INTEREST
The authors declare no conflict of interest.

AUTHOR CONTRIBUTIONS
Samina Kausar and Xuhuahu designed the study, implemented the analysis, and prepared the draft. Samina Kausar also wrote the paper along with literature. Xu Huahu, Shahid Iqbal and rest of the authors contributed in finalizing the manuscript. All authors have read and approved the final manuscript.