Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Adapter für Anomalieerkennung #826

Open
RALFISBORN opened this issue Mar 15, 2023 · 1 comment
Open

Adapter für Anomalieerkennung #826

RALFISBORN opened this issue Mar 15, 2023 · 1 comment

Comments

@RALFISBORN
Copy link

What kind of device or service would you like to see an adapter for?
Add name and company of the device, including links to the device and any additional information[...]
Jeder mögliche Sensor und Aktor, welcher in IOBroker eingebunden ist.

Is the device connected to the internet or only available on a local network?
sowohl als auch

Is an official App or Website available?
nichts bekannt.

Is an official API including documentation available?
nichts bekannt.

Are other libraries for an integration available?
Evtl. als Anregung:
https://www.projectpro.io/article/anomaly-detection-using-machine-learning-in-python-with-example/555

Is this device already integrated in other Smart Home systems?
nichts bekannt.

Is this device already integrated in homebridge? Might the ham adapter in combination with the homebridge plugin be sufficient?
Please try it and add results
nichts bekannt.

Additional context
Die meisten Sensoren sowie Aktoren verhalten sich über die Zeit recht gleich. Interessant wird es gerade für ein Smart Home doch dann, wenn vom Regelverhalten abgewichen wird (Anomalie) und dann eine Aktion ausgelöst wird (z.B. Push Nachricht oder E-mail etc.)
.
grafik

Den Adapter könnte ich mir in zwei Ausbaustufen vorstellen:

  1. Stufe: Im Adapter wähle ich aus welchen Sensor oder Aktor ich überwachen möchte und wie das Normalverhalten ist. Anschließend noch welche Aktion ausgelöst wird.
  2. Stufe: Ich gebe nur noch den Aktor oder Sensor an und gebe einen Zeitraum vor, in dem das Normalverhalten mitgeschrieben wird. Dadurch wird dann das Normalverhalten selber erkannt (z.B. die Haustüre geht pro Tag 5 mal auf). Oftmals ist nicht nur ein Sensor allein für eine Abweichung erforderlich, sondern es sind mehrere, welche in einem Zusammenhang aggieren. Hier wäre es schön, wenn sich über UND/ODER Gatter evtl. Verknüpfungen herstellen lassen. Ein Historie über die Annomalien oder eine Tabelle mit Ranking (Top 5) wäre sicherlich auch hilfreich um Smart Home Systeme eben noch smarter zu machen.

After you create the issue, please vote for yourself in the first post of the issue using the "+1"/"Thumbs up" button

@mcm1957
Copy link
Collaborator

mcm1957 commented Mar 15, 2023

Potenziell related to #524 but for sure a much more complex approach

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants