Data mining uses methods from artificial intelligence in a database context to discover previously unknown patterns. As such, the methods are usually unsupervised. It is closely related but not identical to machine learning. Key tasks of data-mining are cluster analysis, outlier detection and mining ...
1
vote
0answers
16 views
How to proceed if a rule-based classification algorithm finds an instance that can be classified two ways?
I am training a rule based algorithm (PRISM or CN.2) with n classes (y_1,y_2,..,y_n). All rules in the training RuleSet are in ...
0
votes
0answers
21 views
Significance of a 1 state Hidden Markov Model
I've been training different observation sequences to obtain different HMMs corresponding to each observed data. Something intriguing is that I get one observation sequence represented by 1 state. ...
1
vote
0answers
46 views
Why does k-NN perform better than SVR and linear regression?
I have a data set used in a regression with 30 attributes and 30K instances. I am trying out a bunch of algorithms (SMO regression, Linear Regression and K-NN) but it was quite surprising to see that ...
4
votes
1answer
55 views
Using text mining/natural language processing tools for econometrics
I am not sure whether this question is fully appropriate here, if not, please delete.
I am a grad student in economics. For a project which investigates issues in social insurances, I have access to ...
1
vote
3answers
87 views
Skewed data for regression analysis
I have 30 features in my self-collected dataset where I want to build a regression model. When I look at my data, most of the attributes (95% of the data points) are skewed on a very small range. Out ...
1
vote
1answer
52 views
Estimating the probability of next failure?
I have some data about how some components failed. They can either be replaced in whole or have some of their sub-components repaired. I would like to know if there is a way to fit a failure model to ...
0
votes
1answer
24 views
F-measure for document clustering evaluation - NaN
I'm developing the Java application for text document clustering, and I'm researching some evaluation methods. I implemented F-measure (http://en.wikipedia.org/wiki/F1_score), but I have a problem - ...
1
vote
1answer
61 views
Confusion about hidden Markov model
I've gone through Hidden Markov models (HMM) for the past few months. However there are a few things that are confusing.
The set up is simple: I have to model some human gestures such as walking, ...
1
vote
1answer
44 views
Clustering spatio-temporal data?
I have data in the form of timestamp,lat,long which is gps data for users.
I'm new to data mining and want to understand how can I start clustering these data to understand more about it.
Should I ...
0
votes
1answer
56 views
Are there examples of labelled and unlabelled data?
Can someone please provide examples of labelled and unlabelled data? I have been reading definitions of semi supervised learning but it does not make clear on what the two actually are.
0
votes
0answers
35 views
How to build an automated Question/Answer system using machine learning techniques?
I would like to build an automated Q/A system using machine learning concepts.
Would it be optimal to build this system using Bayesian network, simple if-else system, or something else more ...
0
votes
0answers
33 views
How do I check if there is overfitting in weka?
In weka, how do I check if an induced tree overfits the training data?
0
votes
2answers
32 views
Weka GUI - How to compare 2 classifiers after testing trained model?
I'm new to Weka (and machine learning) so this question would be a bit silly.
So I have 2 models built using J48 and RandomForest (both run with 10-fold cross-validation mode) on a 40,000-tuple ...
3
votes
2answers
95 views
Why is k called representer of evaluation in the definition of kernel functions
Why is $k$ called representor of evaluation?
From the book "Learning with kernels" by Bernhard Schölkopf we have the following lines (page 33):
$\langle k(.,x),f\rangle = f(x)$, in particular ...
0
votes
0answers
21 views
Is discretization still the only way to deal with continuous and count variables in data mining association algorithms?
I have recently read a book chapter of data quality in which the author is against turning continuous variables in groups. While I agree with some of his arguments, I was not be able to find a way to ...
0
votes
0answers
67 views
AIC vs BIC vs MDL
I am trying to learn the difference between the three approaches and their applications.
a) As I understand,
AIC = -LL+K
BIC = -LL+(K*logN)/2
Unless I am ...
1
vote
1answer
64 views
Ordinary Least Squares method: why are my regression results insignificant?
I have a problem in my thesis results of OLS regression being insignificant.
I have 3 sectors and each sector has 130 observations.
Is this sample size is sufficient or not ?
Can anyone suggest ...
0
votes
2answers
100 views
How to interpret the output of the summary method for an lm object in R? [duplicate]
I am using sample algae data to understand data mining a bit more. I have used the following commands:
...
0
votes
0answers
30 views
time complexity and space complexity for HMM forward recursion
When Reading the HMM models, I found the following discussion on the time complexity and space complexity regarding forward recursion.
I am sort of confusing on the reason of getting O(K^2N) and ...
0
votes
0answers
40 views
Appropriate method for supervised learning of small data set with few variables
What method exc. for regression can be used in order to get y=f(x1,x2) on a training set of 800 to 2000 samples? y is a whole number <0,15>, x1,x2 are real <0,40>?
I'm interested in prediction ...
1
vote
2answers
121 views
Why do we use k-means instead of other algorithms?
I researched about k-means and these are what I got: k-means is one of the simplest algorithm which uses unsupervised learning method to solve known clustering issues. It works really well with large ...
2
votes
0answers
62 views
SVM classifier (with soft-margin) implementation in R, gamma value and quadprog
I'm trying to implement a Support Vector Machine classifier in R and I have to solve the optimization problem using the quadprog R package which solves problems of the form :
$$min_b \frac{1}{2} ...
0
votes
1answer
39 views
What are new tasks in data mining?
Beside five "classical" data mining tasks
regression and classification
association rules
clustering
outlier detection
dimension reduction and visuzalization,
I recently found sources regarding ...
1
vote
0answers
59 views
Data mining and time series : algorithm suggestion
I'd like to predict the behaviour (next action) of a internet user who had subscribed to a newsletter.
4 actions can be done by the user :
Don't open (/)
Read
Click
Buy
There is a hierarchy in ...
0
votes
0answers
34 views
How can rough sets be applied in data mining?
I read some articles where theory of rough sets is also considered as data mining algorithm. Hovewer, I have not found so far any example when this theory may be useful in solving data mining ...
0
votes
0answers
58 views
Today's popularity of main data mining and machine learning tasks
In my dissertation about clustering, I would like to start with showing how clustering is becoming more and more popular in recent years in comparison with other data mining and machine learning tasks ...
1
vote
0answers
11 views
Iterated Conditional Mode approximation in E step of EM
I wanted to know what is the mathematical justification for using ICM as an approximation for the E step in an EM algorithm.
As I understand in the E step the idea is to find a distribution that is ...
1
vote
0answers
31 views
Using PCA to merge and grade correlated items
I have a real estates' condos sold dataset with the following fields
DOM: Date on the market
sellPct: Percentage difference between the original and final price.
other fields such as Exposure( ...
2
votes
1answer
84 views
How to compare two datasets using metrics drawn from unknown distributions and with small sample sizes?
I have two datasets consisting of metrics from several experiments. Dataset 1 is the collection of results of experiments E performed by user A on product A, repeated N times. Dataset 2 is the ...
1
vote
1answer
33 views
Bias term in support vector machine
In SVM, there is a bias term. But looks to me there are very few discussions on the physical meanings of this term. Why should we have that? How does this term affect the model?
0
votes
0answers
45 views
Binary Classification
I have conducted 2 experiments on two parameters, $P1$ and $P2$, and I want to classify the two parameters for each experiment.
Experiment 1
...
0
votes
0answers
47 views
Statistics and Data Analysis for Small Businesses [closed]
I am currently a forth year mathematics and economics student. Before this I practiced graphic design. To help pay for school I have continued doing design for a small business in my home city. It is ...
0
votes
0answers
14 views
How does the Cross Validation in PRIM work?
One of the steps described in the PRIM algorithm is, after the peeling and pasting procedure, using Cross Validation to select an appropriate Box from the sequence of boxes obtained by the peeling and ...
2
votes
1answer
84 views
What does it imply when an estimate is not inside its 95% confidence interval?
What does it actually imply when a 95% CI does not contain an estimate (coefficient or parameter). Is there some model assumption that has not been satisfied? Or it means something else?
I know when ...
0
votes
0answers
36 views
How does R{MASS} lda function use MLEs to improve its result?
I am using the LDA function in the MASS package of R, which has the following specification:
...
1
vote
2answers
93 views
Highly unbalanced test data set and balanced training data in classification
I have a training set with about 3000 positive instances and 3000 negative instances. But my test data set is pretty much un-balanced. The positive set only has 50 instances and negative has 1500 ...
0
votes
0answers
31 views
Matching curves
I've a large number of curves over a finite interval. Given a new curve C, I want to find a set of curves in the database s.t the area between a matched curve and C is less than a certain threshold.
...
1
vote
0answers
20 views
Hierachical Predictors in a Regression
Note: Mainly this question pertains to predictions from a model.
If the unit of analysis of a regression (or any predictive model really) is the individual retail store and these stores are organized ...
1
vote
1answer
42 views
How to perform Normalization on Call Details Record to perform k-Mean Clustering
I'm new to data mining and currently doing mining project on telecom customer segmentation (based on profile and call details record). I have gender, age, call time and call duration and have to ...
6
votes
2answers
116 views
Is there overfitting in this modellng approach
I recently was told that the process I followed (component of a MS Thesis) could be seen as over-fitting. I am looking to get a better understanding of this and see if others agree.
The objective of ...
4
votes
1answer
111 views
Timeline of machine learning and data mining breakthroughs
Is there any timeline or historical overview of the most important breakthroughs in machine learning and data mining?
0
votes
0answers
27 views
coordinated dual descent method and sequential minimal optimization
Libsvm uses the sequential minimal optimization as its main solver while Liblinear uses coordinated dual descent method. What are the major differences between these two methods? Looks like both of ...
0
votes
0answers
64 views
Rank of within-class scatter matrix in LDA
Let $N$ be the number of total training examples from $C$ classes. Could anyone tell me why the rank of the within-class scatter matrix $S_w=\sum_{i=1}^C(N_i-1)S_i$ (where $S_i$ is the covariance ...
0
votes
0answers
54 views
Variables selection (continuous and classification): how to do in R?
My dataset have both classification (categorical) and continuous variables, ~ 300 variables in all. I'm looking for a way to reduce my attributes to be less than 300 and put them in the decision tree ...
4
votes
1answer
91 views
How should we interpret the variable created by Principal Component Analysis?
I tried to model $\text{Saving} = a + b_1*\text{Income} + b_2*\text{Wealth}$ but found that $\text{Income}$ and $\text{Wealth}$ were highly correlated. I applied PCA to get a new variable $\text{New}$ ...
0
votes
0answers
59 views
How to identify a new pattern in a URL with a machine learning algorithm (Text mining)
I am trying to identify new patterns after analyzing a number of URLs. So let's say, I am investigating the hypothetical website Yoohle.com and their URLs have the following structure.
domain = ...
2
votes
1answer
51 views
Datamining and time series forecasting
Could we say time series forecasting is a part of data-mining or it's just a data-mining tool?
2
votes
3answers
68 views
parallelism in data mining softwares
I'm working on a data set for order prediction/classification and a close deadline upcoming. Fortunately, my university has a super-computer with restricted access. I was thinking of using a few nodes ...
1
vote
1answer
45 views
How can I calculate leaves and nodes of a C 4.5 decision tree?
I have a given amount of attributes, for example 1024, and now want to calculate the amount of leaves and nodes C 4.5 produces (approximately).
Is there something like that for binary trees - a ...
8
votes
3answers
148 views
Detecting clusters in a binary sequence
I have a binary sequence such as 11111011011110101100000000000100101011011111101111100000000000011010100000010000000011101111
Where clusters of mostly 1's are ...
