# Tag Info

14

If you have a neural network (aka a multilayer perceptron) with only an input and an output layer and with no activation function, that is exactly equal to linear regression. In your case, each attribute corresponds to an input node and your network has one output node, which represents the target value you're trying to predict. However, most neural ...

13

I suggest a variation of distribution counting: Read the text and insert all the word encountered into a trie, maintaining in each node a count, how often the word represented by this node has occured. Additionally keep track of the highest word count say maxWordCound . -- $O(n)$ Initialize an array of size maxWordCount. Entry type are lists of strings. -- $... 10 You might be interested in learning about grammar induction: given a set of examples of strings from a context-free language, there are algorithms to learn a context-free grammar that generates those strings. To learn more about it, read the Wikipedia article I linked to, and Inducing a context free grammar, Is there a known method for constructing a ... 9 While both Operations Research and Data Science both cover a large amount of topics and areas, I'll try to give my perspective on what I see as the most representative and mainstream parts of each. As others have pointed out, the bulk of Operations Research is primarily concerned with making decisions. While there are many different ways to determine how ... 8 http://gate.ac.uk/ie/ gives a very nice, concise distinction: Information Extraction is not Information Retrieval: Information Extraction differs from traditional techniques in that it does not recover from a collection a subset of documents which are hopefully relevant to a query, based on key-word searching (perhaps augmented by a thesaurus). ... 7 The idea in matrix factorization is to find the latent variables which connect the input and the output. Suppose that we are interested in a movie recommendation system, and that movies "live" on a one-dimensional axis, having romantic comedies in one end, and action movies in the other. Each "input" and "output" move can be rated in this scale (say$1$to$-...

6

Multiplying by the $n * p$ matrix decreases the dimensionality of the data set. Think of this as projecting the highly dimensional space into a smaller dimensional space. For example, you could do principle component analysis and project it into a small space. This way things that are correlated together are projected into the same dimension and if one of ...

6

It depends entirely on context. In mathematics as a whole, $\log$ usually denotes the natural logarithm (base $\mathrm{e}$). In computer science, the situation isn't as clean because we often want to talk about things like a number of bits or the height of a binary tree and, in those cases, the most natural (*baddum-tsh*) logarithm to use is base-$2$. ...

5

This isn't a full answer, since mhum's is quite good in contrasting the differing aims of OR vs DS. Rather, I want to address this comment of yours: I was wondering if, for example, one could use any OR techniques to solve DS problems. The answer is yes. The clearest example that comes to mind is Support Vector Machines (SVMs). To "fit" an SVM model to ...

4

Check out Metacademy. It is a wonderful learning guide for many topics in machine learning. Specifically, this page lists various resources for learning about support vector machines.

4

One of the popular models is the Bag of Words model Also, you can model the words as integers.. they have 'relative distance metrics' for that, and capture the very essence of the classification process. However a downside to that is the preprocessing step is expensive and also you need to have some domain knowledge. A pretty famous distance metric is the ...

4

Your first step is to characterize what effect you expect an event to have on your signals. Does it change the mean? Increase the mean? Change the variability? The more you can say about the type of effect it will have, the more specific a test you'll be able to build, and thus the more effective any analysis is likely to be. Then, your second step is ...

4

The state of the art in such problems is done these days via deep neural networks. Among others, two popular and recent approaches for solving the problem of detection and localization of objects are the YOLO paper, and the faster-RCNN, which run a classifier over many variously sized regions in an image. As humans, boats and cars are popular object ...

3

You just play once with it, observe it, and you have the model. Try something non obvious, or simply cheat with answers. The game: you think of something, and 20Q asks questions to find what it is. They just collect the answers from the players. They have a collection of questions that are supposed to discriminate solutions. They ask questions to users, ...

3

It's hard for me to tell, but it sounds like you want to automatically detect the mammograms where the outline is incorrect. I'm assuming the desired outline is the perimeter of some convex region in the image. I'm not sure whether clustering is going to be the ideal approach here. The first approach that comes to mind for me is something like this: we ...

3

For a formal definition of document in the Information Retrieval context, you can look at IR glossaries. A common definition is: Document: Specific unit of retrieval (usually text). It can be a web page, an article, a book, a section or chapter. (see for example the Glossary of the book Modern Information Retrieval by Ricardo Baeza-Yates and Berthier ...

3

As far as I know there is no unique definition for what is an outlier/anomaly. Therefore you'll have to decide by yourself what the characteristics of outlying data points in your data set are. This could be for example: distance to the cluster center (threshold), local neighbourhood (a data point that has no/few data points in its neighgbourhood might be an ...

3

Your algorithm does not even run in time $O(n \log n)$; inserting $\Theta(n)$ things in a hashtable costs time $\Omega(n^2)$ already (worst-case). What follows is wrong; I'm leaving it here for the time being for illustrative purposes. The following algorithm runs in worst-case time $O(n)$ (assuming an alphabet $\Sigma$ of constant size), $n$ the number of ...

3

The gathering of occurrence counts is O(n), so the trick is really only finding the top k occurrence counts. A heap is a common way to aggregate the top k values, although other methods can be used (see https://en.wikipedia.org/wiki/Partial_sorting). Assuming k is second param above, and that it's a constant in the problem statement (it appears to be): ...

3

If the size of the alphabet is constant and $n$ denotes the length of the document (i.e. the total number of characters), you can indeed use Tries to get $O(n)$ running time - regardless of the length of the words. In a Trie, you can search for (or insert) a word in time linear in the length of the word. You can use a Trie to keep track of the words as you ...

3

First, the definition is clear: support is exactly "the fraction of transactions that contain a particular subset of items." This is a data-mining term, not a statistics term. $supp(A)$ is the fraction of transactions that contain item $A$. $supp(B)$ is the fraction of transactions that contain item $B$. $supp(A,B)$ is the fraction of transactions that ...

3

I'm not a Data Mining expert, but from what I understand, it means the same thing as it does outside of data mining: to improving the solution in a locally optimal way, rather than one that is guaranteed globally optimal. (And, as Raphael comments, to not backtrack on that decision). For example, if you are training a graphical model, the process of ...

3

Yes, your data is "time-series data", since it's a sequence of measurements of the same variable collected over time. Time-series data can be collected continuously or at discrete intervals. Your sample data can be expressed as a function of time - maybe it helps to think of the "function" as the process that produces the measured output, the input to the ...

3

No. You can't. You need ground truth. You're asking "if I don't know which claims are fraudulent, can an algorithm somehow determine that for me?" The answer of course is no: the algorithm doesn't know anything more than you do about insurance fraud -- if anything, it knows even less. This is one of the challenges with using machine learning. You ...

3

It sounds like you're saying that your approach is able to effectively find the shape of the expression, but it takes it a lot longer to find the right constants. One possible approach is to generate an expression of the correct shape using your genetic programming method, then fill in the constants using a different method. I would recommend trying ...

3

In Grammatical Evolution there are three well known approaches to the problem of constant creation: expression based (the "traditional" approach). This is what you're using (arithmetic operators are required to produce new constants); digit concatenation. An example is: <int> ::= <int><digit> | <digit> <digit> ::= 0 | ...

3

Assuming that your data comes from a Markovian source, you can estimate the entropy of the source using an optimal compression algorithm such as Lempel–Ziv, whose theoretical version (without limiting the table size) is known to asymptotically converge to the entropy. That is, if the entropy of the source (suitable defined) is $H$, then the expected ...

3

Say you have $n$ entries, $m$ distinct letters, and the $i$-th letter occurs $k_i$ times. You can find the last occurrence of the $i$-th letter using a variant of binary search: Double your step size until you find a different letter (or run out of the array), do binary search between your last and second-to-last step. This takes about $2\log k_i$ many steps....

3

Roughly, the algorithm needs to estimate the probability to assign a point the correct cluster. So the algorithm add P to a cluster if it is very unlikely that, after all the points have been processed, some other cluster centroid will be found to be nearer to P. So the algorithm measure the probability that, if P belongs to a cluster, it would be found as ...

3

A metric space consists of a set $X$ of "points" and a metric $d\colon X \times X \to \mathbb{R}_{\geq 0}$ (giving the "distance" between any two points) which satisfies the following constraints: Symmetry: $d(x,y) = d(y,x)$ for all $x,y \in X$. Non-triviality: $d(x,y) = 0$ if and only if $x = y$. Triangle inequality: $d(x,y) \leq d(x,z) + d(z,y)$ for all \$...

Only top voted, non community-wiki answers of a minimum length are eligible