Summary of the JBC090 Lab Sessions and Exercises 2021/2022
30 views 0 purchase
Course
Language and AI (JBC090)
Institution
Technische Universiteit Eindhoven (TUE)
Book
Speech and Language Processing
This summary contains all the theory from the lab sessions and exercises given in the JBC090 course. This is an addition to the summary of the theory given in this course. This will help you prepare for the interim exam as well as the final exam, good luck!
Summary JBC090 Cognitive Science II (Language and AI) 2021/2022
Samenvatting Taaltheorie en Taalverwerking Deeltentamen 2
Samenvatting Taaltheorie en Taalverwerking Deeltentamen 1
All for this textbook (5)
Written for
Technische Universiteit Eindhoven (TUE)
Data Science
Language and AI (JBC090)
All documents for this subject (4)
Seller
Follow
Lieve12
Reviews received
Content preview
Lieve Göbbels
Cognitive Science II (JBC090)
Semester 1, 2021-2022
Cognitive Science II (exercises)
Introduction 2
Distances 2
tf*idf 2
Collecting Data 3
RegEx 3
Minimum Edit Distance 5
Classi cation 6
n-Gram language modeling 6
Regression prediction 8
Fitting regression 8
Evaluating (linear) regression 8
Evaluating classi cation 8
K-Nearest Neighbors 9
Information gain (Decision Trees) 9
Naive Bayes 9
Smoothing across techniques 10
Representation 11
Positive Pointwise Mutual Information (PPMI) 11
Forward propagation in Neural Nets 11
Hidden Markov models (HMM) 14
Recurrent Neural Network Language Model (RNNLM) 18
Lab 1 22
Lab 2 22
Lab 3 23
Lab 4 23
Lab 5 24
, Introduction
In short:
• Distances
• tf*idf
Distances
n
( p i⃗ − q i⃗ )2
∑
Euclidian Distance:
i=1
|A ∩ B|
Jaccard coe cient: J(A, B) =
|A ∪ B|
p⃗∙ q⃗ n
p i⃗ ⋅ q i⃗
∑
Cosine Similarity: where ∙=
p⃗∙ p⃗⋅ q ⃗∙ q ⃗ i=1
Note: p ⃗ ∙ p ⃗ is the ℓ2 norm for vector p .⃗ If one ℓ2 normalizes the full space, the denominator
drops and the cosine similarity becomes: ∥ p ⃗∥2 ∙ ∥ q ⃗∥2
tf*idf
N
wt,d = log(tf(t, d ) + 1) ⋅ logb dft
where tf = term frequency
df = document frequency
N = number of documents
b = base; typically 10 (log = ln; log10 = lg)
The term frequency refers to how many times the term t occurs in a document d. The document
frequency refers to in how many documents the term t occurs. When making a term frequency
matrix, the unique words (the vocabulary; features) are listed in the columns and the documents (or
their numbers, e.g. doc1, doc2) are listed in the rows. The values in the matrix are the counts of each
word in a particular document.
the cat sat on mat my
doc1 (the cat sat on the mat) 2 1 1 1 1 0
doc2 (my cat sat on my cat) 0 2 1 1 0 2
Using the aforementioned formula (wt,d), one can convert this tf matrix to a tf * idf matrix. For
example, wthe,1 = ln(2 + 1) ⋅ lg( 21 ) = 0.331.
The benefits of buying summaries with Stuvia:
Guaranteed quality through customer reviews
Stuvia customers have reviewed more than 700,000 summaries. This how you know that you are buying the best documents.
Quick and easy check-out
You can quickly pay through credit card or Stuvia-credit for the summaries. There is no membership needed.
Focus on what matters
Your fellow students write the study notes themselves, which is why the documents are always reliable and up-to-date. This ensures you quickly get to the core!
Frequently asked questions
What do I get when I buy this document?
You get a PDF, available immediately after your purchase. The purchased document is accessible anytime, anywhere and indefinitely through your profile.
Satisfaction guarantee: how does it work?
Our satisfaction guarantee ensures that you always find a study document that suits you well. You fill out a form, and our customer service team takes care of the rest.
Who am I buying these notes from?
Stuvia is a marketplace, so you are not buying this document from us, but from seller Lieve12. Stuvia facilitates payment to the seller.
Will I be stuck with a subscription?
No, you only buy these notes for $5.07. You're not tied to anything after your purchase.
