Andhra Pradesh reports 92% accuracy for its Vidyarthi Nestham model

A review for UNICEF relays Andhra Pradesh's figures that its dropout prediction model reached 92% accuracy in 2018-19. The reviewers could not establish whether dropout fell.

  • 2 for an organization

    The department reports that the model's accuracy in predicting dropout rose from 75% in 2017-18 to 92% in 2018-19.

  • Not rated a person

    The reviewers could not establish whether the mentoring reduced dropout, because the pilot was only then concluding.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported benefit.
Evidence

foundstudiedreportedclaimed

An independent descriptive report supports the benefit.

The system’s part
Not isolated. The review describes clustering and a boosted decision tree model trained on 2016-17 student data with Azure Machine Learning, so the system learns from data. The accuracy figures are the department's, and the review does not compare dropout with and without the model.
Not yet checked
LimitsCosts and harmsLearns from dataLater outcome

Download data Cite Suggest a correction

Benefit type
Improved Task Accuracy
System
Vidyarthi Nestham dropout prediction model
Made by
Andhra Pradesh School Education department, with support from Microsoft
Used by
Andhra Pradesh School Education department (pilot in 10,000 government schools)
Where
Andhra Pradesh, India
Stage
In a pilot
Date
, when the source was published
Kind
Aggregate result
Broad benefit
Analyzing Data or Information, Identifying Objects, Actions, and Events
Last revised
1 November 2026

Sources

  1. Descriptive report, by authors with no tie to the system.

    Cambridge Education. Review of Education management and monitoring system at school and sub-national level: Report for EMIS in Andhra Pradesh. Prepared for the UNICEF Regional Office for South Asia, 5 August 2020. Published in UNICEF's evaluation database.

    The accuracy of the model progressed to 75% in 2017-18 and then to 92% in 2018-19

Details

AI system

Workflow

The model applied clustering and boosted decision trees to individual student records in the state's SIMS database. It used 64 parameters, with learning outcomes weighted most. The department refined the model each year against actual dropout. Teachers received predicted dropouts to mentor and reported on them every fortnight through a mobile app.

Follow-up

Conditions for replication

Each child must have an individual record in the student information system. Teachers must follow up the students assigned to them and report through the app.

Entry details

Sector

Education.

Purpose

The Andhra Pradesh School Education department built the dropout prediction model, with support from Microsoft, to identify students likely to drop out so that teachers can mentor them.

31 more questions not yet checked

model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). Andhra Pradesh reports 92% accuracy for its Vidyarthi Nestham model (AIBD-1170). Artificial Intelligence Benefits Database, edition of 1 November 2026. https://petervartanian.xyz/aibd/records/AIBD-1170.html (archived at https://doi.org/10.7910/DVN/AADSII, version 7.0)

BibTeX Data Field definitions