Computer-assisted essay assessment
-
Upload
ahmed-cain -
Category
Documents
-
view
30 -
download
2
description
Transcript of Computer-assisted essay assessment
Computer-assisted essay assessment
Textbook paragraphs
………
………
………
………
………
Compute similarity betweentextbook paragraphs and
training essays
………
………
Essays of training m
aterial
………
Define the threshold values forgrade categories
Compute similarity betweentextbook paragraphs and
essay to be graded
Assign essay gradeComputed essay
grade
Essay to be graded
………
Similarity scores by Latent Semantic Analysis
Comparison material based on relevant passages from textbook
Defining threshold values for grade categories
Grading the essays
Results
Results
Training material
Wei
ghti
ngsc
hem
e
Nu
m. o
f gr
aded
essa
ys
Exa
ct
Exa
ct o
rad
jace
nt
Cor
rela
tion
5 sections and 70 essays LE 73 46.6 87.7 0.81
27 paragraphs and 70 essays LE 73 50.7 89.0 0.81
147 sentences and 70 essays LE 73 35.6 91.8 0.79
5 sections and 70 essays LE 73 43.8 87.7 0.80
27 paragraphs and 86 essays LE 57 43.9 84.2 0.80
5 sections and 70 essays N 73 28.8 58.9 0.56
Latent Semantic Analysis (LSA)aka Latent Semantic Indexing (LSI)
Several Applications Information Retrieval Information Filtering Essay Assessment
Documents are presented as a matrix in which each row stands for a unique word and each column stands for a text passage (word-by-document matrix)
Truncated singular value decomposition is used to model latent semantic structure
Resulting semantic space is used for retrieval Can retrieve documents that share no words
with query .
Latent Semantic Analysis (LSA)
Singular Value Decomposition Reduces the dimensionality of word-by-document
matrix Using a reduced dimension new relationships
between words and contexts are induced when reconstructing a close approximation to the original matrix
Reduces irrelevant data and “noise”
Latent Semantic Analysis (LSA)Document comparison
Semantic space is constructed from the training material
To grade an essay, a matrix for the essay document is built
Document vector of essay is compared to the semantic space
doc1 d o c 2 doc3 … d o c n
T1 w11 w12 w13 … w1n
T2 w21 w22 w23 … w2n
T3 w31 w32 w33 … w3n
… … … … …
Q u e ryv e c to r tm wm1 wm2 wm3 … wmn
t1 qw1Similarity scores
t2 qw2 doc1 doc2 doc3 … docn
t3 qw3 S1 S2 S3 … Sn
… …
tm qwm
Compute similarity between documentvectors and query vector
Word-by-document matrix
Latent Semantic Analysis (LSA)
Document comparison Euclidean distance Dot product Cosine measure
Cosine between document vectors
YX
YX
cos
Dot product of vector divided by their lengths
B
A
Latent Semantic Analysis (LSA)
Pros Doesn’t just match on terms, tries to match on
concepts
Cons Computationally expensive, its not cheap to
compute singular values Choice of dimensionality is somewhat arbitrary,
done by experimentation
Latent Semantic Analysis (LSA)
Word-by-document matrix
Latent Semantic Analysis (LSA)
Singular value decomposition
Latent Semantic Analysis (LSA)
Two dimensional reconstruction of word-by-document matrix
Latent Semantic Analysis (LSA)