Questions tagged [semi-supervised-learning]
The semi-supervised-learning tag has no usage guidance.
0
votes
0answers
17 views
Graph-based Label Spreading Implementation
I have a graph with a set of nodes as shown below.
Graph with some nodes labeled belonging to 2 clusters:
Some nodes have a known labels (green or yellow colour) where as the labels of the rest of ...
2
votes
0answers
6 views
Neural Network for detecting/checking for requirements in diagrams
My question is more about what approach is a good/the best approach for my problem:
THE PROBLEM -
I'm an (mechanical/software) engineer and we take extensive amount of time to review technical ...
0
votes
0answers
5 views
Weakly supervised learning and missing labels for data that likely contains that label
Cross-post from Cross Validated, because here seems more approperiate.
I would like to know how to deal with data that misses a label, but is likely to contain the label in a weakly supervised ...
1
vote
0answers
31 views
Improving automated ingestion system using Machine Learning and/or NLP
I'm working on a automated ingestion system which takes a PDF or doc file or a URL. It then parses the file and get me the required text in a json format but there are some error and there are few ...
0
votes
1answer
21 views
Adapting Neural Network to new domain without labels
Is there an approach for the following problem:
Lets say, I trained a neural network on a big dataset for categorizing different fruits in $k$ classes. Afterwards I got a nice model, which performs ...
1
vote
0answers
7 views
formalizing the machine learning model training by incorporating the human involvement
In some learning system designs, sometimes we incorporate a user-interface that enable the users or domain experts to manually change the models by some operations such as, change the rules or some ...
0
votes
0answers
12 views
Self-training in Semi-supervised Learning
I am currently reading notes online on types of semi-supervised learning and one of the simplest techniques mentioned is "self-training". However, one thing I do not understand is that we take a ...
0
votes
0answers
13 views
Semi-supervised learning for regression
It is mentioned on this page that Label Propagation of scikit-learn can be used for regression also. However, nothing is ...
0
votes
1answer
49 views
how to build a predictive model without training data neither historical data
I m trying to score "how much a product is expected in the market". I created some features:
How much this product is used each year.
Where was it used .
how many product for each country.
the main ...
2
votes
1answer
199 views
Semi Supervised Learning without label propagation
I am trying to cluster some words by affinity.
Using Word2Vec I obtained vector representation of every word that I can cluster with a normal unsupervised method.
Of these words, though, I know the ...
1
vote
0answers
26 views
Why should each layer's child network output be close to parent network's output for variance regularizer?
I am reading up on PEA (Pseudo ensemble agreement) regularizer. specificaly in the neural networks domain. It introduces the concept of perturbing the model a little and forcing the model to make ...
0
votes
0answers
20 views
Active learning methods and insight extraction
I am looking for active learning strategies,
I have a data set, I train over it, and now need to use the model on the
usage site itself, there I would like to have a clever strategy for selecting ...
1
vote
0answers
25 views
Semi-supervised Learning doubt
I'm reading "Hands on machine learning" by Aurelien Geron.
He stated that semi-supervised learning is:
Some photo-hosting services, such as Google Photos, are good examples of this. Once
you ...
2
votes
2answers
72 views
Time series binary classificaiton with labelling issues
My situation is quite complicated so I will give a similar example from a simpler domain. Suppose we want to try to predict WHEN a mobile game users will make a purchase if given a sale. Almost every ...
0
votes
0answers
91 views
Machine Learning Algorithm for Dynamic Environments
Which methods are best for managing and predicting and labeling data in dynamic environment? The system data distribution changes and it is not static. The system can have different normal settings ...
0
votes
1answer
72 views
suggest ingredients based on recipe title [closed]
I would like to construct system that would suggest user ingredients once he/she inputs title of the recipe.
I think that this is the task of machine learning or AI, but on the other hand I am pretty ...
0
votes
0answers
17 views
Proper treatment of imperfect labeled data
In typical semi-supervised learning frame, the input data contains both labeled and unlabeled data. But my problem is different with it. In our problem, the "unlabeled" data can not be labeled exactly ...
0
votes
0answers
32 views
Semi-Supervised approaches for unlabeled datasets
I have an unlabeled dataset that I need to run a classification problem on it. I found online some research papers about "Ladder Networks" for unsupervised learning Introduction to Semi-Supervised ...
1
vote
1answer
28 views
Problem faced when collect data randomly from cluster [closed]
I have a semi structured data set. I need to collect some data (unlabeled) randomly for labeling.
As initiative at first I separated labeled and unlabeled data. Then I convert those data from string ...
-1
votes
1answer
217 views
How to do semi-supervised learning for regression
I am doing regression analysis on a data set with over 20000 samples using scikit learn. Trying to use regression models to fit three features to label which is a score ranges from 0 to 10. Problem is ...
0
votes
1answer
28 views
Can You Purposely Bias A Clustering Model?
We have a large amount (Billions) of high cardinality, mixed nominal & numerical data, and are performing some clustering on it as an experiment. There is a small subset of these data, however, ...
2
votes
1answer
4k views
Custom conditional loss function in Keras
I'm looking for a way to create a conditional loss function that looks like this: there is a vector of labels, say l (l has the same length as the input x), then for a given input (y_true, y_pred, l) ...
4
votes
4answers
2k views
Why positive-unlabeled learning?
Machine learning can be divided into several areas: supervised learning, unsupervised learning, semi-supervised learning, learning to rank, recommendation systems, etc, etc.
One such area is PU ...
0
votes
0answers
42 views
Having measurements of a medical instrument how to improve model with demographic data?
There is a medical instrument that produces various measurements.
These measurements are being used as inputs to a model to classify the health condition of a patient. The health condition is binary, ...
1
vote
0answers
264 views
How should I construct a binary classifier for thousands of positive data and millions of unlabeled data?
So far, I have stumbled upon many advices and papers on PU Learning and Unary classification.
TLDR: Does anyone have suggestions for specific algorithm or implementation for labeled data of only one ...
3
votes
1answer
132 views
Probability for label correctness in semi-supervised learning
I am aware of the existence of semi-supervised learning approaches, such as the Ladder Network, where only a subset of the data is labeled. Are there any methods or papers which consider correctness ...
4
votes
2answers
330 views
General strategy for imbalanced, semi-supervised, sparse problem
I am looking for some general advice on where to start with this problem. There are 350 sparse (low positive integer) features. I have 2000 positives, 1000 negatives, and infinite unlabeled data, ...
5
votes
3answers
133 views
Predictive clustering
I have an hypothesis but i don't know if it's true.
If the cluster is dense and we apply a supervised learning on this data, the model generated by this cluster will be more efficient for new data ...
2
votes
1answer
179 views
Generic strategy for object detection
I have a huge collection of objects from which only a tiny fraction are in a class of interest. The collection is initially unlabelled, but labels can be added using an expensive operation (for ...
11
votes
3answers
1k views
Build a binary classifier with only positive and unlabeled data
I have 2 datasets, one with positive instances of what I would like to detect, and one with unlabeled instances. What methods can I use ?
As an example, suppose we want to understand detect spam ...
