Questions tagged [segmentation]

In marketing/economics, the task of dividing some population such as customers into sub-groups based on some type of shared characteristics or demands. Statistically, various unsupervised and supervised methods may be used for that (clustering, conjoint analysis, tree, etc.).

56 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
4
votes
0answers
710 views

BatchNorm after ReLU

I am currently experimenting with different settings for a U-Net (https://lmb.informatik.uni-freiburg.de/people/ronneber/u-net/) based image segmentation and I was unable to find out if it makes any ...
3
votes
0answers
173 views

Potts model and model selection

Suppose to have two vectors of random variables $\mathbf{x}=(x_1,\dots,x_n)^{\prime}$ and $\mathbf{y}=(y_i,\dots,y_n)^{\prime}$, observed over a two dimensional space domain. Let $\boldsymbol{\xi}_{...
3
votes
0answers
102 views

Profiling survey respondents with both categorical and ordinal variables

I have a mixture of categorical and ordinal variables from a survey that I am trying to use to create "profiles" or segments that differ from one another with respect to a dependent variable (the ...
3
votes
1answer
4k views

3-4-5 Rule How to partition the sets?

In Data Mining course, we are taking 3-4-5 Rule to segments the data uniformly. I'm trying to understand these lines below, and how they are linked to the graph below too. If an interval covers 3, ...
2
votes
1answer
36 views

Using pretrained segmantation network for unseen motives

For a research project, I need to do a segmentation on images. Since the motivation is nothing any of the big networks was ever trained on, I would ask if it still makes sense to use pretrained ...
2
votes
0answers
312 views

Evaluate image segmentation with the absence of ground truth

Motivation: - Evaluate the computerised image segmentation against manual segmentation; - Evaluate the difference between difference manual segmentation. Background: Given a raw medical image (...
2
votes
0answers
31 views

What are good / simple techniques available for segmenting non-cursive handwriting images?

I need to process English hand-written form fields. So the hand-writings are expected to be mostly non-cursive but the letters may occasionally overlap with each other slightly, with some punctuation ...
2
votes
0answers
36 views

Clustering stream of new customers based on future potential

I need to cluster new customer according to their future potential, but I have only information about their first transaction. I have access to all transactions for the other customers. So I can do ...
2
votes
1answer
150 views

Clustering Consumer data with over 100 variables and 50000 rows each

I am tasked with performing a clustering exercise for a consumer survey dataset with the hopes of finding distinct consumer segments. In the past, I've done it using a variety of techniques- ...
2
votes
0answers
158 views

What form of analysis should be used for loss given default estimates?

I am working on BASEL II IRB models and we have to estimate loss based on historic defaults. There are different outcomes/scenarios we have identified that a default can encounter that will affect ...
1
vote
0answers
16 views

Find interesting segment automatically in funnel analysis

This is a feature that I came across in Mixpanel and really enjoy it, but I haven't figure out how to do it myself (so I can use it on some other dataset). Use case I have a funnel (which I defined ...
1
vote
0answers
36 views

How can I isolate objects from a 3D image volume using normally distributed geometric features

I have '.tif' image stacks that I am analyzing as volumes. For every object, I can get the Volume, surface area:volume, 'sphericity' and Euler number. For every one of those features, the objects I am ...
1
vote
0answers
140 views

Personalize Recommendations for small dataset

I'm working on a recommender system for a set of niche products. These are products that don't have a large number of customers. Does anyone have any tips on algorithms or approaches that work well ...
1
vote
1answer
48 views

Feature Identification for customer profiling

Problem: I want to identify the characteristic(s) of people who would shop on Monday vs Sunday (or any such dichotomous response variable). I have over a million observations and >50 variables/...
1
vote
0answers
109 views

Hierarchical clustering on dependent data

I do customer clustering based on weekly data. I don't want to rely on only one week of data so I was considering pooling the data over a few weeks together and assume that these are simply "different"...
1
vote
0answers
318 views

supervised binning

I have continuous outcome variable and continuous independent variable. I am trying to bin the independent variable that maximizes homogeneity within bins based on the outcome and maximize separation. ...
1
vote
0answers
37 views

Segmentation - Making Segments Sharper

I have a question. We’ve done a study that looks at the youth’s attitudes towards “social issues” such as water, sanitation and public health. We then segmented the population using factor analysis (...
1
vote
0answers
641 views

Cost function for image segmentation

I am currently working on image segmentation based on superpixels. My input is a data matrix that contains stixels (rectangular superpixels that span an entire column). In the matrix I have stixel ID, ...
1
vote
0answers
58 views

Classifying according to relationship between variables

I do have a large dataset and would like to classify it according to the relationship between some variables. I do not want to use clusters, because clustering results groups that are similar. I do ...
1
vote
0answers
299 views

Interpretation of cross validation results when comparing models

I'm trying to solve a bio-medical image segmentation problem using a binary classifier and then a spatial smoothing (assuming continuous regions). I have: Training set of 10 3D scans, a total of ~30 ...
1
vote
0answers
27 views

Change in distribution of a portfolio of customers

I'm looking for some high-level thoughts on understanding a change in distribution for a portfolio of customers. I have a file with information on a bunch of customers. These customers are purchasing ...
1
vote
0answers
120 views

Goodness of fit of linear model for Segmentation of GPS positions time series

I have some GPS coordinates series taken in regular time steps and I need to verify whether some chunks of the trajectories fit well as a straight line or not. The aim is to perform segmentation on ...
1
vote
0answers
275 views

Test for whether a change is different in direction for subgroups?

I have a (large) population on which I randomly split into groups, and subjected one group to an experimental condition (call it change A) which demonstrated a significant improvement in a target ...
1
vote
0answers
227 views

Best Practices for Using Geodemographic Segmentation Data

I have a data set that clusters block groups in the US into either 15 broad neighborhood categories or 72 fine-grained segments with goofy names. The segments were constructed using factor analysis ...
0
votes
0answers
10 views

Does anyone have any experience in segmenting motor evoked potentials?

I'm working in EEG and motor imagery and was wondering what the best way to go about sectioning my data points for feature analysis / time synchronizing with events or observable features. Are there ...
0
votes
0answers
11 views

Semantic Segmentation Multi-Class Single Channel Output Math

For semantic segmentation problems, I understand that it's a pixel-wise classification problem. At the last layer of the neural network, I would basically have a 1x1x1 convolution layer with a softmax ...
0
votes
0answers
9 views

Semantic segmentation without background class

For the topic of Few-Shot Learning, I am trying to reuse a segmentation model trained on dataset $D_{1}$ with $classes = \{Background, Person\}$ as a pretrained model for the same training set but ...
0
votes
0answers
9 views

About Image Size in image segementation

Objective : A detection aid model for medical doctors to detect brain tumor. sample size : (155, 240, 240) seg size : (155, 240, 240) input shape : (1, 128, 160, 128) output shape : (1, 128, 160, ...
0
votes
0answers
16 views

The optimal number of segments for Piecewise Aggregate Approximation (PAA)

I have time series dataset and I want to break it into segments, in order to run autoencoder on individual segments. After some research, PAA does what I need: it breaks the dataset into segments (...
0
votes
0answers
5 views

If a model can overfit to a minibatch, can I say it can be generalized well with large enough samples?

I am doing a proof-of-concept thing to see if Mask RCNN can do a good instance segmentation on my own dataset. The issue is that I have to annotate the data myself and it takes long time to annotate ...
0
votes
0answers
27 views

Data balancing in image classification

I've to segment defects from an image. The image consists of only tomatoes with it's defects in it. The defects and tomatoes in the dataset are as follows: ...
0
votes
0answers
10 views

Improving bounding box IoU for document segmentation

I have tried using RCNN and faster RCNN to detect tables and paragraphs. However, I keep encountering the same issue, which I guess is because of region suggestion algorithm; Parts of the first or ...
0
votes
0answers
101 views

Multivariate change point detection with strongly correlated (nonlinear) variables

I have a two-variable time series. There is a very strong nonlinear correlation between the two variables, so they can be thought of as: variable_1 = $X$ (random variable) variable_2 = $f(X) + N(\mu,...
0
votes
1answer
16 views

Is the capacity of a multitask U-Net with two-decoders the same of a standard U-Net with doubled capacity in the decoder?

I implemented a U-Net with an additional decoder (one encoder, then it splits into two decoders). The first decoder predicts the normal segmentation label and the second decoder predicts the distance ...
0
votes
0answers
24 views

Looking for classic image segmentation tool for small RGB images (python)

I am working with 32x32 colour images and want to segment items on them. I am using https://scikit-image.org . This only seems to work on grayscale images and mainly seem to focus on detecting edges ...
0
votes
0answers
37 views

Meaning of sparse annotation for images?

what is sparse annotation? Is it pixel-wise labeling for images. what are the other types of annotation?
0
votes
0answers
19 views

Grouping already known objects from an Image

Assuming my typical training data are images that each contain multiple animals, with the animals themselves already located/segmented (We have information for each pixel whether it is an animal or ...
0
votes
0answers
220 views

Does Sørensen–Dice Coefficient (Dice Score) only account for true positives?

I'm working in a project on medical image segmentation which uses the Dice Score as part of the loss function, but I got some doubts with the commonly adopted implementation. The definition of Dice ...
0
votes
0answers
32 views

Krippendorff's Alpha for unitizing texts

I have a question about using alpha to measure Krippendorff's inter-annotator reliability in qualitative data. After reading the papers, I'm still not sure how to handle this case. I'm working on a ...
0
votes
0answers
25 views

what is the best approach in dealing with large dimension custom data for training and predicting deep learning models

i am trying to implement semantic segmentation for satellite images.My custom dataset has dimensions(height,width)in range (3000, 3000)what is the best approach for feeding(for training) and ...
0
votes
0answers
11 views

Approaches for a “risk segmentation” applied to auto insurance

I wonder about approaches for a risk segmentation applied to auto insurance. i.e. Suppose I have all the states within United States, and I wanna say: "Ok, Alaska and Alabama belong to group 1; ...
0
votes
0answers
58 views

Modeling on entire dataset vs. Combining segmentation models trained on subsets of the same dataset

Training machine learning models on an unbalanced dataset: about 3% positive labels, and 97% negative. The modeling goal is to get as many examples as possible with 60% precision on a holdout test set ...
0
votes
2answers
35 views

How to include percentage variables in PCA + K-means when some values are undefined because the denominator is 0?

I'm trying to do customer segmentation by using PCA to reduce dimensionality and then feeding the resulting principal components into a K-means algo to get at the final segments. Some of my variables ...
0
votes
0answers
23 views

Training neural network with a single instance at a time

For a semantic segmentation problem attempted using neural networks, does it make sense to try achieving overfitting with single training example and then (depending on generalization error) add more ...
0
votes
1answer
27 views

How to find categorical contributing factors for an anomaly?

Given a house sales dataset with number of houses sold each day and their attributes (i.e., price, number of rooms, size, etc.) - if on a specific day there's a spike/drop in sales, what are some ...
0
votes
0answers
18 views

How to use Bayesian belief Network map/Causality map for segmentation?

I have obtained the causality map for my data. I have an event of interest and the evidences for the event. How do I make the use of this information to come up with segmentation/clustering such that ...
0
votes
1answer
26 views

What methods could/should I use for identifying sub-groups in a customer segmentation?

I am working on a segmentation model that has been fed into the company by an external agency which created the segmentation based on surveys. I have a number of 6 attitudinal segments and after ...
0
votes
0answers
122 views

Why has the Auxiliary loss fallen out of favor?

Why has the Auxiliary loss fallen out of favor? In deep learning, at one point, there were many deep models that utilize multiple softmax loss so that the gradient can flow better at the beginning of ...
0
votes
1answer
52 views

How we determine the ground truth box of the object in each frame in Matlab?

When we track one object in a video sequence using a tracking object method, the estimated bounding box is given by the method for every frame of the video. But how we determine the ground truth box ...
0
votes
1answer
239 views

Multi Label Semantic Segmentation - One class is way behind

I'm trying to build a Multi-Label Semantic Segmentation model, but while training, when I'm looking at the validation set, I can see that one label is far far behind, and in the end, he is not getting ...