Email Sentiment Analysis with n8n

1. n8n

This guide will show you how to self-host n8n, a workflow automation platform, and use it to create a workflow that automatically analyzes the sentiment of all incoming emails on your account.

This is a completely free process that only requires that you have access to (1) an email account with IMAP/SMTP options and (2) a computer or machine where you can install and use Docker.

1.1. Installation

To get started, read the self-hosted-ai-starter-kit project README or simply clone the repository and set it up with Docker using the command below.

git clone https://github.com/n8n-io/self-hosted-ai-starter-kit.git
cd self-hosted-ai-starter-kit
docker compose --profile cpu up

This will clone the repository, start the stack of Docker containers with compose, and make the n8n web page available at https://localhost:5678. The first run will ask you to configure the administrator account for n8n.

1.2. Reverse Proxy

If you want to use n8n from a public domain name, you'll need to configure a reverse proxy. I use Nginx on my server for reverse proxies. Nginx configuration locations and practices vary by distribution, but the following commands will show you my general reverse proxy configuration.

cd /etc/nginx/conf.d
nano n8n.conf

server {
	listen                  443 ssl;
	listen                  [::]:443 ssl;
	http2			on;
	server_name             n8n.example.com;
	# SSL
	ssl_certificate         /etc/letsencrypt/live/example.com/fullchain.pem;
	ssl_certificate_key     /etc/letsencrypt/live/example.com/privkey.pem;
	ssl_trusted_certificate /etc/letsencrypt/live/example.com/chain.pem;
	# reverse proxy
	location / {
		set $upstream_n8n http://127.0.0.1:5678;
        proxy_set_header Remote-User $user;
		proxy_set_header Remote-Email $email;
		proxy_set_header Remote-Groups $groups;
		client_body_buffer_size 128k;
		proxy_next_upstream error timeout invalid_header http_500 http_502 http_503;
		send_timeout 5m;
		proxy_read_timeout 360;
		proxy_send_timeout 360;
		proxy_connect_timeout 360;
		proxy_set_header Host $host;
		proxy_set_header Upgrade $http_upgrade;
		proxy_set_header Connection upgrade;
		proxy_set_header Accept-Encoding gzip;
		proxy_set_header X-Real-IP $remote_addr;
		proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
		proxy_set_header X-Forwarded-Proto $scheme;
		proxy_set_header X-Forwarded-Host $http_host;
		proxy_set_header X-Forwarded-Uri $request_uri;
		proxy_set_header X-Forwarded-Ssl on;
		proxy_redirect  http://  $scheme://;
		proxy_http_version 1.1;
		proxy_set_header Connection "";
		proxy_cache_bypass $cookie_session;
		proxy_no_cache $cookie_session;
		proxy_buffers 64 256k;
    }
# HTTP redirect
server {
	listen      80;
	listen      [::]:80;
	server_name n8n.example.com;
	include     custom.d/letsencrypt.conf;
	if ($host ~ ^[^.]+\.example\.com$) {
		return 301 https://$host$request_uri;
	}
}

Simply restart Nginx to ensure the new configuration is in-effect.

sudo systemctl restart nginx.service

1.3. Workflow Configuration

You can open the included workflow at http://localhost:5678/workflow/srOnR8PAY3u4RSwb or simply open the web interface and create a new configuration.

This workflow contains the following nodes:

1. Email Trigger (IMAP)
   • Create an IMAP credential with your email's IMAP settings to allow n8n to monitor for new emails.
2. If
   • Check {{ $json.subject }} to see if it contains "n8n Sentiment Analysis" to ensure we don't re-check our own emails.
3. Sentiment Analysis
   • Analyze the {{ $json.textPlain }} input coming from each email.
   • Ensure you have Include Detailed Results enabled so that we can access the confidence and strength variables.
4. Ollama Chat Model
   • The Docker stack we used includes Ollama and Llama3, which provide the easiest way to test this workflow.
5. Send Email
   • Create an SMTP credential to allow n8n to send emails.
   • Subject: n8n Sentiment Analysis: {{ $json.sentimentAnalysis.category }}
   • Email Format: Text
   • Text (Expression): This can contain anything you want, but be sure to include the variables {{ $json.sentimentAnalysis.category }}, {{ $json.sentimentAnalysis.strength }}, and {{ $json.sentimentAnalysis.confidence }}.
   • In the additional options, I enabled the Append n8n Attribution option in the screenshots below.

1.4. Testing

You can use the Test Workflow button at the bottom of the canvas area to test the workflow. This relies on receiving new messages in your inbox, so be sure to send yourself a test email!

1.5. Results

After testing each step noted above, n8n provided the results below - it works!

While this isn't anything earth-shattering, it does show easy it is to get started with n8n and large language models in a self-hosted environment.

...or, comment on this post on Bubbles!