week1_allquestions.txt

?L1.5: 5'20"-3'50": #C# How many dimensions does one term in a VSM define? (A) 3 (B) 2 (C) 1 (D) 4
?#A# Which of the following best describes a Bag of Words? (A) An N-Dimensional Vector space (B) Scrabble Bag (C) Collection of terms relevant to bags
?how to define syntactic structures?


week2_allquestions.txt
?I still dont understand how to compute Gamma code, and why it is better than binary.
?L2.2: 6'03"-6'39":
?L.4: 4'59" - 5'06": Why is inverted index the most common type of indexing used?
? 1107 - 1616 : Professor talks about gamma-code. Is there way to directly compute log of x rather than recognizing unary code and getting the value of log of x by plugging the unary code formula ? 

week3_allquestions.txt
?When would we use binary judgements? Why wouldnt we just use multi level judgements as it allows for more flexibility?
?L3.2 Is this F-measure similar to F-statistic in statistics?


week4_allquestions.txt
?L4.6-4.7 Is there one smoothing function better or more common than the rest?
?Why not have a standard smoothing language model? Whats the point of having multiple, if at the end of the day, they still assign probabilities while including unseen words?

week5_allquestions.txt
?Do all search engines only use implicit feedback system as its the most convienient for users while giving semi-reliable results.
?Generic Mixture Model>Kullback-Leibler Divergence Retrieval Model>Link Analysis=HIPS=PageRank>Rocchie Feedback>Web Indexing

week6_allquestions.txt
?How do you effectively set up a filtering system with no bias.
?659-830: how should we decide the value of beta? Should the beta we choose has a greater influence on alpha than gamma and N do? Or in other words, overall, should beta control alpha or N and gamma control alpha?

week7_allquestions.txt
?Are there any other applications for Logic predicates being used in text representation?
?13'14" - 13'45": How IDF(w) is defined?
?Are all words that are not in a paradigmatic relation in a syntagmatic relation? 

week8_allquestions.txt
?Probabilistic Topic Models>Overview of Statistical Language Models=Topic Mining and Analysis>Mutual Information>Syntagmatic Relation Discovery: Entropy & Conditional Entropy
?I don't understand what ML estimation does.

week9_allquestions.txt
?#A# In the expectation algorithm, do we seek to minimize or maximize theta? (A) Maximize (B) Minimize     
?What is the point of z in PLSA

week10_allquestions.txt
?#A# True or False: in Top-Down Clustering, we gradually partition the data into smaller clusters. (A) True (B) False    
?Should you first determine the class a text belongs to and then cluster?
?difference between generative probabilistic model for cluster and categorization?

week11_allquestions.txt
?#B# True or False: In SVM classifiers, we assume beta1 > 0 and beta 2 < 0 (A) True (B) False
?If most of this class is based around ML, can't we use elements of AI to boost our applicability of algorithms?
?L11.2: 17'30"-'": Is C the covariance matrix? Or just a regularization parameter, and is it tuned by trial and error?
?What is an example of perform error analysis to obtain insights to design effective features?
? #B# In the sentence "In response to the damage of Hurricane Katrina, the governor issued a speech.", the "the governor" is the target, and  "Hurricane Katrina" is the holder. (A) True, (B) False

week12_allquestions.txt
?#A# True or False: Text can be associated with nodes of a network and subnetworks. (A) True (B) False