0
votes
0answers
33 views

How to perform proper data mining on time-series data?

I have some daily data from city A, B, C. Values from city A are highly correlated with values from other cities for lag -1,-2,-3 and -4. I want to use Random Forest, SVM and ANN to predict values ...
0
votes
0answers
24 views

In which Data Stream Mining Algorithms do Damped Windows make sense?

For Data Stream Mining, especially in Document Classification, the most common ML algorithms are Multinomial Naive Bayes, Stochastic Gradient Descent and Ozbag (ADWIN). When looking at their ageing ...
0
votes
0answers
25 views

Calculate similarity of waiting times of users

Let's say I have waiting times(seconds) of users in web pages. ...
4
votes
2answers
315 views

Data mining techniques in R for advertising and sales data

I would like to apply one or more data mining techniques to a dataset, in order to see the effect advertising has on sales. I am working from this dataset. It has 36 consecutive entries of monthly ...
0
votes
0answers
28 views

Literature for prediction models where each training example has a different amount of data?

This could be a machine learning question as much as a statistics question, but I think this is the best place to put the question. Here are three different examples of problems where each ...
0
votes
1answer
51 views

How to find which time series is trending more?

Let us say I have two sets of time varying series as shown below: ...
1
vote
0answers
32 views

How to find the most important factors or combination of levels in a finite data set

I’d like to discover which “calendar factors” (e.g. day of the week, month) have the strongest relationship with whether or not a particular product is sold at least once. For days when the product ...
0
votes
0answers
45 views

Data mining of time series

I have a dataset which consist of n time series variables $X_1$..$X_n$ , and a time serie output $Y$. I would like to mine the data to find if some functions (lagged or not) of the $X_i$ can predict ...
0
votes
0answers
56 views

Time Period Predictive Modeling

I have been implementing some classification algorithms (Naive Bayes, SVM etc) recently on the iris data sets to get head start into the data science field. I enjoy working on machine learning ...
0
votes
0answers
45 views

Gather insights from quarterly financial forecast data

I am trying to analyze a quite large (~25,000 rows) dataset of financial forecasts. The forecasts are usually not derived from algorithms, but come from a large number of analysts who forecast the ...
5
votes
2answers
219 views

Ordinal/continuous vs dummy variable for time series regression/data mining

Let's suppose I have a time series data that I would like to regress $y$ on $x$ and $Time$. See below for the dataset. ...
1
vote
1answer
66 views

The concept of “average run length” in change point detection

With respect to the change point detection for data stream, there is a concept of "average run length", which is discussed in the CPM package manual: I am not clear why that "average number of ...
5
votes
2answers
183 views

What to do with my data?

I'm quite new here so please excuse me if this isn't a suitable question. I'm developing a site for students to allow them to upload their virtual transcripts (grade report) to our site and allow us ...
0
votes
0answers
31 views

Comparing Twitter Trends on a Monthly Basis?

I'm trying to compare the frequency of a keyword is mentioned on Twitter in a monthly basis (e.g.: how many times the keyword "MissWorld" appears on September, how many it appears on August, how many ...
0
votes
0answers
60 views

Finding motif in an on-line (real time) time series

I have a data stream (one integer value per second): 480, 481, 479, 482,... If I show these values in a 2D-diagram, I have something like (for limited values): ...
0
votes
0answers
44 views

Transform a daily closed values to data of rolling annual returns.

My problem is that : I have the daily closed values of the initial index for DJUSER, MSCI, SP500, SPGSCI from 1 January 1999 to 31 December 2011. I want to transform them in to data of rolling annual ...
1
vote
0answers
128 views

Features selection by filter methods for multivariate time series

I have a data set in which the samples are multivariate (about 30 variable/features) time series. These samples refer to two classes. I would like to select the variables more relevant to discriminate ...
0
votes
0answers
49 views

Help with Algorithm Selection

I have asked this on the rapid-miner forum (http://forum.rapid-i.com/) but it can't hurt to get other opinions due to it not being a software-specific question. If this is the wrong section please let ...
1
vote
1answer
120 views

How to improve forecasting accuracy?

I got some users' history data and generated some sequences of real numbers. The length of each sequence is between 15 and 25. What's more, I do not know whether these sequences have patterns and the ...
0
votes
0answers
198 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 ...
2
votes
1answer
64 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?
0
votes
2answers
621 views

Time series prediction - what is Autoregressive Tree model ? (Python)

Our problem: model evolution of values of a continuous variable over time. I came through a paper presenting an approach for predicting the next values for a time series. Whereas ARIMA model is more ...
3
votes
1answer
57 views

Methods for teasing apart the influence of different time series features on a target feature?

Are there any established methods for teasing apart the influence of different time series features on a target feature? To illustrate: The target: Sales volume of kittens. Features: Time of year, ...
2
votes
1answer
213 views

Modelling longitudinal data

We have longitudinal data on children(n<20) in which we measure different quantities A,B,C,D (like distance walked, time spent in school etc.). These are all continuous variables. We measure these ...
3
votes
1answer
51 views

What is the difference between emerging and discriminative patterns?

I'm working on a way to make efficient classification on sequential patterns mining algorithms. Looking in the state-of-the-art I found that there is two type of algorithm to address this problem : ...
1
vote
2answers
493 views

Temporal association rules with arulesSequences with quantities

I am trying to mine product-usage sequences for multiple users of online gaming site. I have found the R package arulesSequences but am not sure how to fit it to my problem. The data format would be ...
1
vote
1answer
553 views

Clustering of time series

I have a set of almost 1600 time series on 2 years which I want to group into clusters. Do you think this is possible using k-means? Which method do you advice me to use? Is this possible at all using ...
6
votes
1answer
264 views

Finding the correct data mining approach

(I apologise for being a newb, but I'm a researcher introducing myself to data mining---any help or insight would be greatly appreciated. Also, this isn't technically a homework question, but I've ...
1
vote
0answers
121 views

Modern and practical methods for forecasting time series cross section data

I have a data array $X$ which is of size $T \times N \times K$, where $T = 1500$, $N = 1500$ and $K = 10$. Physically, the 1st index $1, 2, \ldots, T$ denotes days, while the 2nd index $1, 2, ...
3
votes
3answers
570 views

Univariate clustering of time series

I just want to know if its possible to cluster an univariate time series, in order , say, to detect anomalies? and do you have any online version for denstream code, in Matlab? here is the time ...
4
votes
2answers
867 views

Pre-processing time series data for data mining / predictive modeling input

What are some ways to prepare/pre-process time series data to use the series data as a predictor(s) in a predictive model (classification or regression)? Specifically, what are the methods to be ...
2
votes
1answer
195 views

Multiple values in a timestamp

I am working with a time series (discrete) that has ideally 1 value per time stamp. In some cases, we are seeing multiples that have a wide range all recorded with the same time stamp. Up to now, we ...
5
votes
3answers
2k views

Outliers spotting in time series analysis, should I pre-process data or not?

My question builds on a previous post on outlier detection in generic time series, and specifically on the answer provided by the always great Rob H. I work for a small-sized manufacturing company ...
2
votes
1answer
160 views

Time series digitization

I am looking for a viable method to create an alphabet out of multidimensional time series data. My intention is to use pattern/string matching algorithms (potentially genetic) to find profitable ...
4
votes
1answer
1k views

Best practices for measuring and avoiding overfitting?

I am developing automated trading systems for the stock market. The big challenge has been overfitting. Can your recommend some resources describing methods for measuring and avoiding overfitting? I ...
2
votes
0answers
70 views

Determine optimal (recurring) frequency of discrete values

So I'm trying to determine a semi-efficient way to calculate the optimal (recurring) frequency of a set of data. The data only exists at random time periods, but it is assumed that the data set is ...
0
votes
2answers
193 views

Ways of finding associations in time series

I am trying to develop a logic to identify association between different time series for association mining. I have a lot of series and need to find whether or not the association exists. I figured ...
2
votes
1answer
2k views

Sliding window validation for time series

I have a broad question about sliding window validation. Specifically, I am looking at using Rapid Miner to predict future values of a financial series using "lagged" values of that series and other ...
40
votes
3answers
1k views

Do we have a problem of “pity upvotes”?

I know, this may sound like it is off-topic, but hear me out. At Stack Overflow and here we get votes on posts, this is all stored in a tabular form. E.g.: post id voter id vote type ...
10
votes
6answers
523 views

Dubious use of signal processing principles to identify a trend

I am proposing to try and find a trend in some very noisy long term data. The data is basically weekly measurements of something which moved about 5mm over a period of about 8 months. The data is to ...