I want to compare the results of two papers, namely the matrix factorization techniques of Koren (2009) and Rendle et al (2009)
The first one uses RMSE values and the second one AUC. Is there any way to compare these?
More specifically, I want to compare the RMSE for implicit feedback of the first paper(which is around 0.89) to the BRP-MF line on the last page of the second paper(with an AUC of 0.91-0.92).
You cannot compare these results at all, for at least 3 reasons:
different metrics
different datasets -- the reported results are not from experiments on the same dataset
different tasks -- rating prediction is different from item recommendation -- you would also not say that predicting the price of something and detecting spam are the same tasks, wouldn't you?
Why would you want to compare such different methods anyway?
Maybe you want to re-read the first paper -- it uses implicit feedback as an additional signal in order to compute better rating predictions, whereas in the second paper implicit feedback is used to predict implicit feedback.
Related
I am rleatively new to statistics and am stuggling with the normality assumption.
I understand that parametric tests are underpinned by the assumption that the data is normally distributed, but there seems to be lots of papers and articles providing conflicting information.
Some articles say that independant variables need to be normally disrbiuted and this may require a transformation (log, SQRT etc.). Others says that in linear modelling there are no assumptions about any linear the distribution of the independent variables.
I am trying to create a multiple regression model to predict highest pain scores on hospital admissions:
DV: numeric pain scores (0-no pain -> 5 intense pain)(discrete- dependant variable).
IVs: age (continuous), weight (continuous), sex (nominal), depreviation status (ordinal), race (nominal).
Can someone help clear up the following for me?
Before fitting a model, do I need to check the whether my independant variables are normally distributed? If so, why? Does this only apply to continuous variables (e.g. age and weight in my model)?
If age is positively skewed, would a transformation (e.g. log, SQRT) be appropriate and why? Is it best to do this before or after fitting a model? I assume I am trying to get close to a linear relationship between my DV and IV.
As part of the SPSS outputs it provides plots of the standardised residuals against predicted values and also normal P-P plots of standardised residuals. Are these tests all that is needed to check the normality assumption after fitting a model?
Many Thanks in advance!
I’m trying to check the performance of my LDA model using a confusion matrix but I have no clue what to do. I’m hoping someone can maybe just point my in the right direction.
So I ran an LDA model on a corpus filled with short documents. I then calculated the average vector of each document and then proceeded with calculating cosine similarities.
How would I now get a confusion matrix? Please note that I am very new to the world of NLP. If there is some other/better way of checking the performance of this model please let me know.
What is your model supposed to be doing? And how is it testable?
In your question you haven't described your testable assessment of the model the results of which would be represented in a confusion matrix.
A confusion matrix helps you represent and explore the different types of "accuracy" of a predictive system such as a classifier. It requires your system to make a choice (e.g. yes/no, or multi-label classifier) and you must use known test data to be able to score it against how the system should have chosen. Then you count these results in the matrix as one of the combination of possibilities, e.g. for binary choices there's two wrong and two correct.
For example, if your cosine similarities are trying to predict if a document is in the same "category" as another, and you do know the real answers, then you can score them all as to whether they were predicted correctly or wrongly.
The four possibilities for a binary choice are:
Positive prediction vs. positive actual = True Positive (correct)
Negative prediction vs. negative actual = True Negative (correct)
Positive prediction vs. negative actual = False Positive (wrong)
Negative prediction vs. positive actual = False Negative (wrong)
It's more complicated in a multi-label system as there are more combinations, but the correct/wrong outcome is similar.
About "accuracy".
There are many kinds of ways to measure how well the system performs, so it's worth reading up on this before choosing the way to score the system. The term "accuracy" means something specific in this field, and is sometimes confused with the general usage of the word.
How you would use a confusion matrix.
The confusion matrix sums (of total TP, FP, TN, FN) can fed into some simple equations which give you, these performance ratings (which are referred to by different names in different fields):
sensitivity, d' (dee-prime), recall, hit rate, or true positive rate (TPR)
specificity, selectivity or true negative rate (TNR)
precision or positive predictive value (PPV)
negative predictive value (NPV)
miss rate or false negative rate (FNR)
fall-out or false positive rate (FPR)
false discovery rate (FDR)
false omission rate (FOR)
Accuracy
F Score
So you can see that Accuracy is a specific thing, but it may not be what you think of when you say "accuracy"! The last two are more complex combinations of measure. The F Score is perhaps the most robust of these, as it's tuneable to represent your requirements by combining a mix of other metrics.
I found this wikipedia article most useful and helped understand why sometimes is best to choose one metric over the other for your application (e.g. whether missing trues is worse than missing falses). There are a group of linked articles on the same topic, from different perspectives e.g. this one about search.
This is a simpler reference I found myself returning to: http://www2.cs.uregina.ca/~dbd/cs831/notes/confusion_matrix/confusion_matrix.html
This is about sensitivity, more from a science statistical view with links to ROC charts which are related to confusion matrices, and also useful for visualising and assessing performance: https://en.wikipedia.org/wiki/Sensitivity_index
This article is more specific to using these in machine learning, and goes into more detail: https://www.cs.cornell.edu/courses/cs578/2003fa/performance_measures.pdf
So in summary confusion matrices are one of many tools to assess the performance of a system, but you need to define the right measure first.
Real world example
I worked through this process recently in a project I worked on where the point was to find all of few relevant documents from a large set (using cosine distances like yours). This was like a recommendation engine driven by manual labelling rather than an initial search query.
I drew up a list of goals with a stakeholder in their own terms from the project domain perspective, then tried to translate or map these goals into performance metrics and statistical terms. You can see it's not just a simple choice! The hugely imbalanced nature of our data set skewed the choice of metric as some assume balanced data or else they will give you misleading results.
Hopefully this example will help you move forward.
I am using Spark ML to optimise a Naive Bayes multi-class classifier.
I have about 300 categories and I am classifying text documents.
The training set is balanced enough and there is about 300 training examples for each category.
All looks good and the classifier is working with acceptable precision on unseen documents. But what I am noticing that when classifying a new document, very often, the classifier assigns a high probability to one of the categories (the prediction probability is almost equal to 1), while the other categories receive very low probabilities (close to zero).
What are the possible reasons for this?
I would like to add that in SPARK ML there is something called "raw prediction" and when I look at it, I can see negative numbers but they have more or less comparable magnitude, so even the category with the high probability has comparable raw prediction score, but I am finding difficulties in interpreting this scores.
Lets start with a very informal description of Naive Bayes classifier. If C is a set of all classes and d is a document and xi are the features, Naive Bayes returns:
Since P(d) is the same for all classes we can simplify this to
where
Since we assume that features are conditionally independent (that is why it is naive) we can further simplify this (with Laplace correction to avoid zeros) to:
Problem with this expression is that in any non-trivial case it is numerically equal to zero. To avoid we use following property:
and replace initial condition with:
These are the values you get as the raw probabilities. Since each element is negative (logarithm of the value in (0, 1]) a whole expression has negative value as well. As you discovered by yourself these values are further normalized so the maximum value is equal to 1 and divided by the sum of the normalized values
It is important to note that while values you get are not strictly P(c|d) they preserve all important properties. The order and ratios are exactly (ignoring possible numerical issues) the same. If none other class gets prediction close to one it means that, given the evidence, it is a very strong prediction. So it is actually something you want to see.
I have training data that falls into two classes, let's say Yes and No. The data represents three tasks, easy, medium and difficult. A person performs these tasks and is classified into one of the two classes as a result. Each task is classified independently and then the results are combined. I am using 3 independently trained SVM classifiers and then voting on the final result.
I am looking to provide a measure of confidence or probability associated with each classification. LIBSVM can provide a probability estimate along with the classification for each task (easy, medium and difficult, say Pe, Pm and Pd) but I am unsure of how best to combine these into an overall estimate for the final classification of the person (let's call it Pp).
My attempts so far have been along the lines of a simple average:
Pp = (Pe + Pm + Pd) / 3
An Inverse-variance weighted average (since each task is repeated a few times and sample variance (VARe, VARm and VARd) can be calculated - in which case Pe would be a simple average of all the easy samples):
Pp = (Pe/VARe + Pm/VARm + Pd/VARd) / (( 1/VARe ) + ( 1/VARm ) + ( 1/VARd ))
Or a multiplication (under the assumption that these events are independent, which I am unsure of since the underlying tasks are related):
Pp = Pe * Pm * Pd
The multiplication would provide a very low number, so it's unclear how to interpret that as an overall probability when the results of the voting are very clear.
Would any of these three options be the best or is there some other method / detail I'm overlooking?
Based on your comment, I will make the following suggestion. If you need to do this as an SVM (and because, as you say, you get better performance when you do it this way), take the output from your intermediate classifiers and feed them as features to your final classifier. Even better, switch to a multi-layer Neural Net where your inputs represent inputs to the intermediates, the (first) hidden layer represents outputs to the intermediate problem, and subsequent layer(s) represent the final decision you want. This way you get the benefit of an intermediate layer, but its output is optimised to help with the final prediction rather than for accuracy in its own right (which I assume you don't really care about).
The correct generative model for these tests likely looks something like the following:
Generate an intelligence/competence score i
For each test t: generate pass/fail according to p_t(pass | i)
This is simplified, but I think it should illustrate tht you have a latent variable i on which these tests depend (and there's also structure between them, since presumably p_easy(pass|i) > p_medium(pass|i) > p_hard(pass|i); you could potentially model this as a logistic regression with a continuous 'hardness' feature). I suspect what you're asking about is a way to do inference on some thresholding function of i, but you want to do it in a classification way rather than as a probabilistic model. That's fine, but without explicitly encoding the latent variable and the structure between the tests it's going to be hard (and no average of the probabilities will account for the missing structure).
I hope that helps---if I've made assumptions that aren't justified, please feel free to correct.
My data set consist of two disjoint subsets. There are two different classifiers with confidence measure, each can work only on one definite subset. I need to enhance accuracy of my system on the whole data, so I need to "combine" confidences of these classifiers. The point is that scales of classifiers can be sharply different, e.g. at one confidence level error and accept can be much different. Maybe there is a way to transform confidences to some uniform scale for both classifiers?
Accuracy means maximization of accept level subject to fixed error rate (for example, 10% of data set)
what about to make 3rd classifier which would make use of your 2 classifiers(input) and its output should be what you expect
How about combining classifiers through boosting. The paper titled "Boosting Localized Classifiers in Heterogeneous Databases" paper gives more detail.