The Sentiment Analysis Tool
Score one text or many for polarity and subjectivity — with optional detailed emotion and aspect analysis — so an agent can quantify how customers, employees, or markets feel, at scale and on infrastructure you control.
Thousands of opinions, no way to read them all
Reviews, survey responses, support tickets, and social posts carry the voice of your customers and employees — but at volume, nobody can read them all, and "the vibe seems negative" isn’t something you can act on.
Volume defeats reading
No one can read thousands of comments to find the trend.
Gut feel isn’t data
"It feels negative" can’t be tracked or compared.
Aspects get lost
Overall sentiment hides what specifically people love or hate.
Sensitive feedback
Customer and employee text can’t be sent to a hosted service.
Feelings, quantified
Scoring
Polarity and subjectivity
Turn opinion into numbers.
The tool scores text for polarity (positive to negative) and subjectivity, so an agent can quantify sentiment across a single message or a whole batch and track it over time.
- Polarity scoring
- Subjectivity scoring
- Single or batch input
- Choice of engine
Positive ↔ negative
Depth
Emotions and aspects
Not just positive or negative.
With detailed analysis enabled, it surfaces emotions and the specific aspects driving sentiment — so an agent learns not just that feedback is negative, but what about it is.
The why behind it
Governance
On-premise scoring
Feedback stays internal.
Scoring can run on a local engine inside your perimeter with audit logging, so sensitive customer and employee feedback is analyzed without leaving your environment.
Private, logged
Parameters
The sentiment_analyze tool accepts these inputs when an agent calls it. Required inputs are flagged.
default: textblob Optional Analysis engine to use. textblobopenai
default: false Optional Include detailed analysis (emotions, aspects).
Where sentiment analysis pays back
Feedback triage
Score support tickets to surface unhappy customers.
Survey analysis
Quantify open-ended survey responses at scale.
Brand monitoring
Track sentiment in mentions over time.
Employee pulse
Gauge sentiment in internal feedback.
Product insight
Find the aspects customers love or hate.
Agent prioritization
Let a support agent escalate by sentiment.
Assigned to agents, orchestrated as networks
On VDF AI, an industry’s use cases map to agents, and you assign tools like this one to those agents. Compose multiple agents into a governed, on-premise network.
What changes after you assign it
Questions about the Sentiment Analysis tool
What does the sentiment analysis tool do?
It scores text for polarity and subjectivity, with optional detailed emotion and aspect analysis. Assigned to an agent, it quantifies how customers, employees, or markets feel across a single message or a large batch.
Can it analyze many texts at once?
Yes. Pass a single text or an array of texts to score a whole batch in one call.
What does detailed mode add?
It surfaces emotions and the specific aspects driving sentiment, so you learn not just that feedback is negative but what about it is.
Is feedback kept private?
Yes. Scoring can run on a local engine on-premise with audit logging, so sensitive customer and employee text never leaves your environment.
How is it used by agents?
Sales, support, and strategy agents use it to triage and prioritize, often alongside the CSV analyzer and document generator for reporting.
Assign Sentiment Analysis to these agents
These VDF AI agents can be assigned this tool. Open an agent to see the full toolkit it can run.
Tools that work well alongside this one
Where this tool delivers value
Quantify how your customers feel
See the sentiment analysis tool let an agent score feedback at scale — on infrastructure you control.