SCHOLARLY PERFORMANCE PREDICTION USING MACHINE LEARNING AND DATA MINING TECHNIQUES

Abstract

Information technology has simplified and enhanced fact series. Without further investigation, this record will remain prevalent. Research and development in various sectors has sought significant insights from these large datasets to solve complex problems. Through innovative methods, information analysis has become highly positive. The computing fields of data mining and machine learning offer fantastic ways to explore large datasets. This article describes several educational programs that leverage those fields to predict student achievement. Early identification of students' performance trends helps improve learning outcomes, where system learning and information mining work best. It also covers Educational Data Mining, a type of statistics mining used in teaching. EDM analyzes teaching and learning statistics using device mastering, statistics, statistics mining, and data analysis. This innovation emphasizes reading systems and fact-gathering to forecast pupils' educational performance. The benefits and usage of certain tactics in training are examined. 5 Claims &1 Figure

Specification

Description:Field of the Invention
The present invention focuses on exploring facts mining algorithms to enhance college students' instructional performance. Its primary intention is to become aware of and reduce the failure fee by implementing well timed interventions to boost scholar consequences. Anticipating college students' performance proves tremendous for both educators and learners, facilitating upgrades inside the teaching and learning approaches.
Objectives of the Invention
The invention's major goal is to early identity of students' overall performance tendencies proves helpful in taking proactive measures to beautify studying outcomes, where system learning, and information mining strategies show best.Furthermore, the thing delves into the specialized use of statistics mining in training, known as Educational Data Mining (EDM). EDM employs various methods from device mastering, statistics, statistics mining, and data analysis to scrutinize statistics amassed for the duration of the teaching and getting to know techniques. And also to emphasize the importance of reading systems and extracting facts to predict students' educational overall performance. It explores how those strategies are used and their advantages inside the discipline of training.
Background of the Invention
The threat sensing system described in (US2018/0158305A1) includes multiple threat sensing devices strategically located throughout a school or facility. Each device consists of one or more acoustic sensors, one or more gas sensors, and a communication circuit or device designed to transmit sensor data to a system gateway. The system gateway is designed to receive and analyze sensor data from threat detection devices to ascertain if the processed data aligns with any of a predefined set of known threats (e.g., a gunshot). If a match is found, it communicates the threat's existence, the processed sensor data, and/or specified messaging information to one or more recipient devices.

A method and system for detecting impaired driving, monitoring, and preventing accidents based on driving patterns (US2021/11161519B2). A technique for conducting an impairment test is shown, which assesses in real-time if a vehicle operator is impaired (e.g., due to drug or alcohol use, distraction, conversation, texting, eating, or drowsiness). The impairment test evaluates multiple assigned probability impairment values over a specified time frame to compare the frequency of received event signals from both the vehicle's passenger compartment and the external environment regarding one or more driver performance actions executed by the vehicle operator. This comparison is made against data previously stored in a unique driver profile to statistically ascertain whether the driver's operation of the vehicle may indicate impairment, based on the distinctive behavior patterns recorded in the profile. The distinctive driver profile encompasses certain behavioral patterns of the driver, including "habit evidence," to assess potential impairment.
A technique for regulating vehicle systems (US2019/0276034A) involves acquiring monitoring data from one or more surveillance systems and ascertaining multiple driving states based on the monitoring data from the aforementioned systems. The approach involves ascertaining a composite driver state derived from several driver states and adjusting the control of one or more vehicle systems according to the composite driver state.

In (US2020/0118164A1), The embodiments disclose an integrated mobile device management method including using an integrated mobile device management service provider digital programmable server and database server for coordinating processing with device issuer locked devices with functionalities specifically targeted users to limit users access to specific functions, coordinating locked devices functionalities for recording and analyzing user information, user device usage and sorting user profiles into layered categories, analyzing data and controlling function processes using at least one customized processor with an embedded algorithm within the integrated mobile device management service provider digital programmable server, downloading locked device functions to at least one locked electronic device for device issuer distribution to targeted users, using the integrated mobile device management service provider digital programmable server for operating an advertiser ad placement auction website, selecting targeted advertisements based on user device usage analysis results, and displaying targeted advertisements on at least one locked electronic devices.

This disclosure pertains to methods, computer program products, and computer systems for a robotic kitchen hub designed for the adjustment and calibration of mini manipulation libraries in multi-functional robotic platforms intended for commercial and residential settings, utilizing artificial intelligence and machine learning. The versatile robotic platform has a robotic kitchen designed for calibration using either a joint state trajectory or a coordinate system, such as Cartesian coordinates, for the mass installation of robotic kitchens. Calibration verifications and adaptations of the micro manipulation library, along with adjustments for any serial or diverse models, provide scalability in the mass production of a robotic kitchen system. A robotic kitchen with multi-mode functionality offers a robot mode for autonomous food preparation, a collaboration mode for task-sharing between the robot and the user, and a user mode where the robot assists the user in preparing a dish.

Automated inventory management systems and methods for medicines and healthcare products stored in bins within care facilities are described. A method entails providing an interactive storage device for attachment to a bin, outputting a visual representation of the bin's local inventory through an audiovisual element, receiving user input, determining modifications to the local inventory based on the user input, updating the local inventory in a non-volatile data store according to the modifications, synchronizing the local inventory with one or more nodes via a communication interface, and receiving periodic updates from the nodes via the communication interface for a local cache that includes locations and inventories of one or more remote bins, as detailed in (US2020/0410446A1).

Methods and apparatus for cleansing a central venous catheter port are detailed in (US2020/0289811A1). An apparatus comprises a body, a coupling designed to attach the body to the hub, a cleaning cap affixed to the body, and an actuator located within the body for rotating and translating the cap in relation to the hub. The cleaning cap comprises a cap body that delineates a cavity, inside which is situated a cleaning member including threads that interlock with the threads on the hub.

Summary of the Invention
The proposed invention will be helpful for the students.It aims to spotlight various packages of those techniques across distinct educational domains. Its primary objective is to underscore the significant ability of instructional information mining tools, strategies, and their pivotal role in forecasting college students' instructional achievement.

Detailed Description of the Invention
It is no longer possible to manually analyze the massive volumes of data that educational institutions collect and store because they have become too extensive. The term "Educational Data Mining" (EDM) refers to a type of information mining that focuses on uncovering hidden insights within student data in order to improve academic performance. According to Iie-Haiyan, Biac, and Yuan (2017), EDM is considered to be an essential component of the training apparatus since it enables remarkable interactions between a large number of factors, with the ultimate objective of improving educational methods. According to Silva and Fonseca (2017), the major objective is to improve academic etiquette and provide insights for the purpose of making intelligent decisions among students.
In its most elementary form, EDM makes use of conventional data mining techniques in order to conduct an analysis of educational data in order to find solutions to pedagogical problems (Baker &Yacef, 2009). The development of e-learning systems (Lara, Lizcano, Martinez, Pazos, and Riera, 2014), the clustering of educational data (Chakrabarty, Chakma, and Mukherjee, 2016), and the prediction of student performance (Chauhan, Shah, Karn, and Dalal, 2019) are only some of the applications that might be made use of this technology. In the field of electronic data management (EDM), some of the most prominent approaches include sequential sample analysis, clustering, prediction, type, machine learning models, and association rule analysis.
In the context of Educational Data Mining (EDM), there are a number of procedures that are engaged; nevertheless, the main process is centered on four primary stages. Following the completion of the Problem Definition Phase, a particular problem is transformed into a data mining challenge, which includes the project's objectives, aims, and major research queries. The Data Preparation and Gathering Stage is an essential component of the analytical technique. It takes up a considerable amount of time, which could be as much as 80 percent of the whole duration of the study. Because of the considerable problem that the data presents, it is necessary to identify, clean, and format the source data into the formats that have been established. Following the completion of this process, the Modeling and Evaluation Phase will begin. During this phase, appropriate parameters will be defined, and a variety of modeling approaches will be selected and carried out. Graphs and reports are used to illustrate the findings of the data mining process, which is organized in a methodical manner during the Deployment Phase.
Students in higher education can benefit from the ability of an organization to forecast results since it can assist them in evaluating their overall performance in a course and enabling them to take appropriate actions. When students are able to anticipate their potential performance, they are able to proactively address any capability constraints that may arise.
The early detection of the performance of college students is important for the implementation of proactive measures to improve the outcomes of learning. Within the realm of educational data mining, one of the most prevalent applications is the prediction of a student's academic progress simply based on past data. This information is a significant resource that may be utilized to enhance the outcomes of students, as stated by Buenaño-Fernández, Gil, and Luján-Mora (2019).
The purpose of this chapter is to investigate a number of applications of data mining (DM) and machine learning (ML) techniques that are utilized to forecast the general achievement of college students. The creation of predictive models can be accomplished through the utilization of a wide variety of methods that fall under the categories of data mining (DM) and machine learning (ML). It is possible that classification is the method that is used the most frequently among these. Neural Networks, Decision Trees, Naïve Bayes, Support Vector Machines, and K-Nearest Neighbors are some of the classification algorithms that are often used for this purpose, as stated by Shahiri and Husain (2015).
5Claims &1 Figure
Brief description of Drawing
In the figure which are illustrate exemplary embodiments of the invention.
Figure 1, the proposed architecture , Claims:The scope of the invention is defined by the following claims:

Claim:
1. A system/method to scholarly performance prediction using machine learning and data mining techniques, said system/method comprising the steps of:
a) The system starts up collected students data (1), then preprocessing will apply to this data to remove the noisy data (2).
b) The developed system will take the inputs and processed with various Algorithms in the model (3), based on the analysis, the system will predict the output (3).
2. As mentioned in claim 1, the system for predicting scholarly performance, comprising a machine learning module configured to analyze academic data inputs and generate performance predictions based on historical patterns derived from data mining techniques.
3. As mentioned in claim 1, the data inputs include academic records, attendance, study habits, and other relevant metrics, which are processed by data mining algorithms to uncover patterns indicative of scholarly performance.
4. According to claim 1, the machine learning module employs supervised learning algorithms to train a predictive model that improves accuracy based on past data and user feedback.
5. As per claim 1, the machine learning model continuously updates its predictions by integrating new academic data, adapting to recent trends and changes in student performance over time.

Documents