Difference between revisions of "Education Pathways"

From Glitchdata
Jump to navigation Jump to search
 
(5 intermediate revisions by 2 users not shown)
Line 1: Line 1:
 +
Looking at data science products in the educational space.
  
 
+
* Get input data
* Badging
+
** Dataset of all courses
* Create dataset of all courses
+
*** Create components of courses
** Create components of courses
+
*** Course tagging, MDM, metadata
** Course tagging, MDM, metadata
+
** Get Student Data
 +
*** Aptitude
 +
*** Location of residence
 +
*** Socio-Economic profile
 +
*** Age etc..
 +
*** Aspirations
 +
** Get environmental data
 +
*** Access to finance
 +
*** Economic conditions
 +
*** Benefits of Educational Outcome / Sentiment Media
 
* Overlay any dependencies
 
* Overlay any dependencies
** Automated identification of dependencies via NLP, downstream word co-relation
+
** Automated identification of dependencies via NLP, topic modelling, downstream word co-relation
 
** Tokenisation, Topic Modelling
 
** Tokenisation, Topic Modelling
** Graph Pathways
+
** Correlation of topics -> Correlation of Courses
 +
** Generate Educational Graph Pathways
 +
** Analyse Graph nodes for waypoints, course clusters, [[Badging]]
 +
*** Allows us to identify duplication in course content
 
* Overlay value of courses
 
* Overlay value of courses
 
** Based on demand, Based on market need, Social feedback
 
** Based on demand, Based on market need, Social feedback
 
** Scoring of courses using sentiment analysis
 
** Scoring of courses using sentiment analysis
 
* [[Recommendation]]
 
* [[Recommendation]]
** Based on user selection?
+
** Based on user selection/user goals?
 
** Based on university marketing?
 
** Based on university marketing?
 
** Based on end goals?
 
** Based on end goals?
Line 19: Line 32:
 
** Recommendation using inputs + trained recommendation model
 
** Recommendation using inputs + trained recommendation model
  
 
+
==Related==
 +
* [[Pathing]]
  
 
[[Category: Data Science Applications]]
 
[[Category: Data Science Applications]]

Latest revision as of 22:19, 16 February 2019

Looking at data science products in the educational space.

  • Get input data
    • Dataset of all courses
      • Create components of courses
      • Course tagging, MDM, metadata
    • Get Student Data
      • Aptitude
      • Location of residence
      • Socio-Economic profile
      • Age etc..
      • Aspirations
    • Get environmental data
      • Access to finance
      • Economic conditions
      • Benefits of Educational Outcome / Sentiment Media
  • Overlay any dependencies
    • Automated identification of dependencies via NLP, topic modelling, downstream word co-relation
    • Tokenisation, Topic Modelling
    • Correlation of topics -> Correlation of Courses
    • Generate Educational Graph Pathways
    • Analyse Graph nodes for waypoints, course clusters, Badging
      • Allows us to identify duplication in course content
  • Overlay value of courses
    • Based on demand, Based on market need, Social feedback
    • Scoring of courses using sentiment analysis
  • Recommendation
    • Based on user selection/user goals?
    • Based on university marketing?
    • Based on end goals?
    • Based on user aptitude?
    • Recommendation using inputs + trained recommendation model

Related