Semantic comparison using MiniLM Sentence Transformer
Copy the two sentences from any example below into Engineering Text A and Engineering Text B, then click Compare Meaning.
Example 1 — Similar Grounding Problem
Text A:
Ground G17 became unsuitable after the ECU was relocated.
Text B:
Moving the control unit increased the distance to the
existing grounding point.
Example 2 — Similar Routing Problem
Text A:
Wiring harness routing is colliding with the body panel.
Text B:
The cable path interferes with the BIW and requires
rerouting.
Example 3 — Similar Clip Problem
Text A:
The edge clip is difficult to remove from the BIW.
Text B:
Excessive retention force makes the harness fixing clip
difficult to detach from the body panel.
Example 4 — Similar Connector Problem
Text A:
Connector locking tab was damaged during assembly.
Text B:
The connector locking mechanism broke during vehicle
installation.
Example 5 — Different Engineering Problems
Text A:
Ground point current exceeded the permitted limit.
Text B:
Connector locking tab was damaged during assembly.
Example 6 — Same Meaning, Different Words
Text A:
ECU relocation increased the harness length to the
existing ground.
Text B:
Moving the electronic control unit resulted in a longer
electrical return path.
Example 7 — Manufacturing Similarity
Text A:
Harness installation is difficult because accessibility
is limited.
Text B:
Assembly operators cannot easily access the wiring
installation location.
Example 8 — Completely Different Topics
Text A:
Splice location needs to be moved closer to the harness
branch.
Text B:
Engineering drawing revision must be released before the
next milestone.
The MiniLM AI model converts Engineering Text A and Engineering Text B into numerical semantic embeddings.
It then compares the two embeddings using cosine similarity.
This means the AI compares the meaning of the sentences, rather than simply checking whether the same words are present.
For example:
Text A:
ECU position changed.
Text B:
Control unit was relocated.
Although the wording is different, the AI may identify them as semantically related because ECU / control unit and changed position / relocated represent similar concepts.
Model: all-MiniLM-L6-v2
Purpose: Sentence embeddings and semantic similarity
Platform: Hugging Face
Runtime: Transformers.js
Processing: Browser
The displayed similarity percentage is derived from cosine similarity between the two embeddings.
It should not be interpreted as a probability that the two engineering statements are identical or technically equivalent.