faculty performance analysis based on student feedback, Slides of Machine Learning

Download faculty performance analysis based on student feedback and more Slides Machine Learning in PDF only on Docsity!

FACULTY PERFORMANCE ANALYSIS

BASED ON STUDENT FEEDBACK

NAME:-NUNNA S K V S S PRAMOD

N B V SUBBA RAIDU

REG.NO:-

36110856

ABSTRACT

• (^) Students feedback is crucial for academic institutions inorder to evaluate

faculty performance.

• (^) Handling the qualitative opinions of students efficiently while automatic

report generation is a challenging task. Indeed, most organizations deal

with quantitative feedback effectively, whereas qualitative feedback is

either processed manually or ignored altogether.

• (^) This study proposes a supervised aspect based opinion mining system

based on two-layered LSTM model. The first layer predicts the aspects

described within the feedback and later specifies the orientation

(positive, negative, and neutral) of those predicted aspects.

• OBJECTIVE

To perform aspect based sentiment analysis on students’

feedback for evaluating faculty teaching performance by using

deeplearning approach.

EXISTING SYSTEM

There is a need to automate this process to analyze students’

feedback and this task comes under the emergent area of opinion

mining. Although, there have been some research studies that attempted

to solve this problem by using opinion mining but their work is confined

to just finding the polarity of an overall sentence.

• (^) PROPOSED SYSTEM

The system performs ABSA(Aspect Based Sentiment Analysis) on student’s

textual feedback to evaluate the teaching quality of a concerned faculty

member.

 Our proposed framework is comprised of two layered LSTM model for aspect

extraction and sentiment classification.

 The first layer classifies a review sentence in one of the six aspects including

TeachingPedagogy,Behavior,Knowledge,Assessment,Experience,and

General.Next,the second LSTM layer predicts the sentiment orientation (+ve,

−ve or Neutral) expressed towards that particular aspect.

Aspect Extraction

It can be defined as the task of identifying an entity’s relevant features from

the opinionated text.Every entity has some sort of features or aspects

associated, for which the opinions are being formed. Most research studies

were purely attempted to extract only aspects from the available text without

classifying text orientation. They performed this task of aspect extraction

through various supervised, semi-supervised, and deep learning approaches.

Their study was based on the assumption that aspects are basically Nouns

and Noun Phrases. Extraction of such noun phrases is being done through the

association rule mining technique.

• (^) Architecture

• ALGORITHM USED

Naive Bayes algorithm that is based on conditional probabilities. Ranking

uses Bayes theory concepts.

A Bayes theorem is a mathematical formula that calculates probability by

including the frequency of values and combinations of values in the data

center.

We would like to give ranking for popular items based on a decision tree. If a

learning algorithm produces accurate class probability estimates, it certainly

produces an accurate ranking. But the opposite is not true

SCREENSHOTS