I'm going to assume that the labels you are creating are for classification or regression. You can consider semi-automated labeling, where the manually annotated labels are extrapolated to new observations that are similar. Here's a simple workflow you can follow:

1. Reduce the dimensionality of your labelled and not labelled data with methods like PCA. plot the first two components as a scatter plot.

2. Use a clustering algorithm like k-means clustering to find the amount of clusters that you have manually annotated observations. For example, if you have 100 manually annotated observations, you find 100 clusters. Now, every observation has a cluster they belong to.

3. Extrapolate the manually annotated labels to other observations inside that cluster.

Keep in mind that this method is highly susceptible to the quality of your manual annotations, and the number of PCA components selected for clustering. Remember that these are weak labels deducted from your manually annotated data, so if the variability of your data is high, then they might not contain enough information for proper imputation.

There are several tools and platforms available to help you efficiently annotate sensor data for machine learning applications. You can explore these options further to find the one that best suits your needs:

• Segments.ai
• Super Annotate
• SensiML Data Studio
• AI-Assisted Data Labeling Platforms
• WebLabel

Remember to select a tool that matches your specific sensor data type, project requirements, and scalability needs.

Dear Rardchawadee Silapunt ,

Data is the currency of the future.

With technology and AI slowly seeping into our everyday lives, data and its proper use can cause a significant impact in modern society.

Accurately annotated data can be used effectively by ML algorithms to detect problems and propose workable solutions, thus making data annotation an integral part of this change.

https://www.v7labs.com/blog/data-labeling-guide

Regards,

Shafagat