This is my first data science project and I need to select some data. Of course, I know that I can not just select all the data available because this will result in overfitting. I am currently investigating house prices in the capital of Denmark for the past 10 years and I wanted to know which type of houses I should select in my data:
Owner-occupied flats and houses (This gives a dataset of 50000 elements)
Or just owner-occupied flats (this gives a dataset of 43000 elements)
So as you can see there are a lot more owner-occupied flats sold in the capital of Denmark. My opinion is that I should select just the owner-occupied flats because then I have the "same" kind of elements in my data and still have 43000 elements.
Also, there are a lot higher taxes involved if you own a house rather than owning an owner-occupied flat. This might affect the price of the house and skew the data a little bit.
I have seen a few projects where both owner-occupied flats and houses are selected for the data and the conclusion was overfitting, so that is what I am looking to avoid.
This is an classic example of over-fitting due to lack of data or insufficient data.
Let me example the selection process to resolve this kind of problem. I will example using the example of credit card fraud then relate that with your question or any future problem of prediction with classified data.
In ideal world credit card fraud are not that common. So, if you look at the real data you will find only 2% data which resulted in fraud. So, if you train a model with this datasets it would be biased as you don't have normal distribution of the class (i.e fraud and none fraud transaction in your case its Owner-occupied flats and houses). There are 4 a way to tackle this issue.
Let's Suppose Datasets has 90 none fraud data points and 10 fraud data points.
1. Under sampling majority class
In this we just select 10 data points from 90 and train model with 10:10 so distribution is normalised (In your case using only 7000 of 43000 flats). This is not ideal way as we would be throughout a huge amount of data.
2. Over sampling minority class by duplication
In this we duplicate the 10 data points to make it 90 data point distribution is normalised (In your case duplicating 7000 house data to make it 43000 i.e equal to that of flat). While this work there is a better way.
3. Over sampling minority class by SMOTE (recommended)
Synthetic Minority Over-sampling Technique is a technique we use k nearest neigbors algo to generate the minority class in your case the housing data. There is module named imbalanced-learn (here) which can be use to implement this.
4. Ensemble Method
In this method you divide your data into multiple datasets to make it balance for example dividing 90 into 9 sets so that each set can have 10 fraud class data and 10 none fraud class data. In your case diving 43000 in batch of 7000. After that training each one separately and using majority vote mechanism to predict.
So now I have created the following diagram. The green line shows the price per square meter of owner occupied flats and the red line shows price per square meter of houses (all prices in DKK). I was wondering if there is imbalanced classification? The maximum deviation of the prices is atmost 10% (see for example 2018). Is 10% enough to say that the data is biased and hence therefore is imbalanced classified?
Related
If you had data like (prices and market-cap are not real)
Date Stock Close Market-cap GDP
15.4.2010 Apple 7.74 1.03 ...
15.4.2010 VW 50.03 0.8 ...
15.5.2010 Apple 7.80 1.04 ...
15.5.2010 VW 52.04 0.82 ...
where Close is the y you want to predict and Market-cap and GDP are your x-variables, would you also include Stock in your model as another independent variable as it could for example be that price building for Apple works than differently than for VW.
If yes, how would you do it? My idea is to assign 0 to Apple and 1 to VW in the column Stock.
You first need to identify what exactly are you trying to predict. As it stands, you have longitudinal data such that you have multiple measurements from the same company over a period of time.
Are you trying to predict the close price based on market cap + GDP?
Or are you trying to predict the future close price based on previous close price measurements?
You could stratify based on company name, but it really depends on what you are trying to achieve. What is the question you are trying to answer ?
You may also want to take the following considerations into account:
close prices measured at different times on the same company are correlated with each other.
correlations between two measurements soon after each other will be better than correlations between two measurements far apart in time.
There are four assumptions associated with a linear regression model:
Linearity: The relationship between X and the mean of Y is linear.
Homoscedasticity: The variance of residual is the same for any value of X.
Independence: Observations are independent of each other.
Normality: For any fixed value of X, Y is normally distributed.
I am new to time-series machine learning and have a, perhaps, trivial question.
I would like like to forecast the temperature for a particular region. I could train a model using the hourly data points from the first 6 days of the week and then evaluate its performance on the final day. Therefore the training set would have 144 data points (6*24) and the test set would have 24 data points (24*1). Likewise, I can train a new model for regions B-Z and evaluate each of their individual performances. My question is, can you train a SINGLE model for the predictions across multiple different regions? So the region label should be an input of course since that will effect the temperature evolution.
Can you train a single model that forecasts for multiple trajectories rather than just one? Also, what might be a good metric for evaluating its performance? I was going to use mean absolute error but maybe a correlation is better?
Yes you can train with multiple series of data from different region the question that you ask is an ultimate goal of deep learning by create a 1 model to do every things, predict every region correctly and so on. However, if you want to generalize your model that much you normally need a really huge model, I'm talking about 100M++ parameter and to train that data you also need tons of Data maybe couple TB or PB, so you also need a super powerful computer to train that thing something like GOOGLE data center. Coming to your next question, the metric, you may use just simple RMS error or mean absolute error will work fine.
Here is what you need to focus Training Data, there is no super model that take garbage and turn it in to gold, same thing here garbage in garbage out. You need a pretty good datasets that can represent whole environment of what u are trying to solve. For example, you want to create model to predict that if you hammer a glass will it break, so you have maybe 10 data for each type of glass and all of them break when u hammer it. so, you train the model and it just predict break every single time, then you try to predict with a bulletproof glass and it does not break, so your model is wrong. Therefore, you need a whole data of different type of glass then your model maybe predict it correctly. Then compare this to your 144 data points, I'm pretty sure it won't work for your case.
Therefore, I would say yes you can build that 1 model fits all but there is a huge price to pay.
I'm applying LSTM on time series forecasting with 20 lags. Suppose that we have two cases. The first one just using five lags and the second one (like my case) is using 20 lags. Is it correct that for the second case we need more units compared to the former one? If yes, how can we support this idea? I have 2000 samples for training the model, so this is the main limitation for increasing number of units here.
It is very difficult to give an exact answer as the relationship between timesteps and number of hidden units is not an exact science. For example, following factors can affect the number of units required.
Short term memory problem vs long-term memory problem
If your problem can be solved with relatively less memory (i.e. requires to remember only a few time steps) you wouldn't get much benefit from adding more neurons while increasing the number of steps.
The amount of data
If you don't have enough data for the model to learn from (which I feel like you will run into with 2000 data points - but I could be wrong), then increasing the number of timesteps won't help you much.
The type of model you use
Depending on the type of model you use (e.g. LSTM / GRU ) you might get different results (this is not always true but can happen for certain problems)
I'm sure there are other factors out there, but these are few that came to my mind.
Proving more units give better results while having more time steps (if true)
That should be relatively easy as you can try few different options,
5 lags with 10 / 20 / 50 hidden units
20 lags with 10 / 20 / 50 hidden units
And if you get better performance (e.g. lower MSE) with 20 lags problem than 5 lags problem (when you use 50 units), then you have gotten your point across. And you can reinforce your claims by showing results with different types of models (e.g. LSTMs vs GRUs).
Consider a problem whose solution maximizes an objective function.
Problem : From 500 elements, 15 needs to be selected (candidate solution), Value of Objective function depends on the pairwise relationships between the elements in a candidate solution and some more.
The steps for solving such a problem is described here:
1. Generate a set of candidate solutions in guided random manner(population) //not purely random the direction is given to generate the population
2. Evaluating the objective function for current population
3. If the current_best_solution exceeds the global_best_solution, then replace the global_best with current_best
4. Repeat steps 1,2,3 for N (arbitrary number) times
where size of population and N are smaller (approx 50)
After N iterations it returns a candidate solution stored in global_best_solution
Is this the description of a well-known algorithm?
If it is, what is the name of that algorithm or if not under which category these type of algorithms fit?
What you have sounds like you are just fishing. Note that you might as well get rid of steps 3 and 4 since running the loop 100 times would be the same as doing it once with an initial population 100 times as large.
If you think of the objective function as a random variable which is a function of random decision variables then what you are doing would e.g. give you something in the 99.9th percentile with very high probability -- but there is no limit to how far the optimum might be from the 99.9th percentile.
To illustrate the difficulty, consider the following sort of Travelling Salesman Problem. Imagine two clusters of points A and B, each of which has 100 points. Within the clusters, each point is arbitrarily close to every other point (e.g. 0.0000001). But -- between the clusters the distance is say 1,000,000. The optimal tour would clearly have length 2,000,000 (+ a negligible amount). A random tour is just a random permutation of those 200 decision points. Getting an optimal or near optimal tour would be akin to shuffling a deck of 200 cards with 100 read and 100 black and having all of the red cards in the deck in a block (counting blocks that "wrap around") -- vanishingly unlikely (It can be calculated as 99 * 100! * 100! / 200! = 1.09 x 10^-57). Even if you generate quadrillions of tours it is overwhelmingly likely that each of those tours would be off by millions. This is a min problem, but it is also easy to come up with max problems where it is vanishingly unlikely that you will get a near-optimal solution by purely random settings of the decision variables.
This is an extreme example, but it is enough to show that purely random fishing for a solution isn't very reliable. It would make more sense to use evolutionary algorithms or other heuristics such as simulated annealing or tabu search.
why do you work with a population if the members of that population do not interact ?
what you have there is random search.
if you add mutation it looks like an Evolution Strategy: https://en.wikipedia.org/wiki/Evolution_strategy
I have a set of 300,000 records of historic customer purchases data. I have started SSAS data mining project to identify best customers.
The split of data:
-90% non-buyers
-10% buyers
I have used various various algorithms of SSAS (decision trees and neural networks showed best lift) to explore my data.
The goal of the project is to identify/score customers according who is most likely to buy a product.
Currently, I have used all my records for this purpose. It feels that something is missing in the project. I am reading two books now about data mining. Both of them talk about splitting data mining into different sets; however, none of them explain HOW to actually split them.
I believe I need to split may records into 3 sets and re-run the ssas algorithms.
Main questions:
How do I split data into training, validation and test sets
1.1 What ratio of buyers and non-buyers should be in a training set?
How do I score my customers according to most likely to buy a product and least likely to buy a product.
The division of your set could be done randomly as your data set is big and the number of buyers is not too low (10%). However, if you want to be sure that your sets are representative you could take 80% of your buyers samples and 80% of non buyers samples and mix them to build a training set that contains 80% of your total data set and it has the same ratio of buyers-non buyers as the original data set which makes the subsets representative. You may want to divide your dataset not in two subsets but in three: training, crossvalidation and test. If you use a neural networkas you said you should use the crossvalidation subset to tune your model (weight decay, learning rate, momentum...).
Regarding your your second question you could use a neural network as you said and take the output, that will be in the range [0, 1] if you use a sigmoid as the activation function in the output layer, as the probability. I would also recommend you to take a look to collaborative filtering for this task because it would help you to know which products may be a customer interested in using your knowledge of other buyers with similar preferences.