# Clementine ADI — oppiOS | OPPI

> The confidence-scoring engine behind OPPI's predictions. Clementine learns from your venue's trading history to forecast outcomes before you commit.

Canonical URL: https://oppi.au/oppios/clementine
Site index for agents: https://oppi.au/oppios/llms.txt

---
[oppiOS · all modules](https://oppi.au/oppios) MOD · clementine

## CLEMENTINE

Collective Learning Engine Managing Efficiency Navigating Tasks And Intelligence in Networked Environments

The ML-powered confidence scoring system that helps OPPI predict outcomes. Named after one of our founders from Concept Cafes, Clementine learns from your venue's historical data to generate confidence ratings for every recommendation.

Capabilities

1.  01

    #### Machine Learning Core

    Continuously learns from your venue's outcomes to improve prediction accuracy over time.

2.  02

    #### Historical Data Analysis

    Analyses patterns from past decisions, seasonal trends, and operational results.

3.  03

    #### Confidence Scoring

    Generates 0-100% confidence ratings on how positive an outcome will be.

4.  04

    #### Effect Prediction

    Estimates the potential positive impact of recommended actions on your venue.

5.  05

    #### Adaptive Learning

    Updates predictions based on real-world results and team feedback.

6.  06

    #### Networked Intelligence

    Learns from anonymized patterns across the OPPI network while protecting your data.

In practice

1.  01

    #### Staffing Confidence

    CCR predicts whether adding a shift will generate positive ROI based on historical patterns.

2.  02

    #### Inventory Decisions

    Confidence scores on reorder suggestions based on consumption trends and upcoming events.

3.  03

    #### Risk Assessment

    Lower CCR flags decisions that need human review due to unusual circumstances.

4.  04

    #### Opportunity Detection

    High CCR highlights actions with strong historical success indicators.

5.  05

    #### Continuous Improvement

    Track CCR accuracy over time to see how OPPI learns your venue's patterns.

oppiOS / clementine

### CLEMENTINE

Capabilities

06

In practice

05

Route

/oppios/clementine

Status

Active

Get started

### Run CLEMENTINE on your venue.

[Get early access](https://oppi.au/getOPPI#signup) [All modules](https://oppi.au/oppios)

CLEMENTINE · oppiOS module

### Who Uses OPPI Hub

Configure once at the top, experience everywhere below. Each role interacts with OPPI Hub differently.

#### Owners & Senior Managers

Trust the engine

##### Review CCR Analytics

Monitor how confident OPPI has been and how accurate those predictions turned out

##### Set CCR Thresholds

Define what confidence level is required before OPPI can auto-implement actions

##### Network Insights

See how your venue's CCR compares to anonymized industry benchmarks

##### Learning Progress

Track how Clementine's accuracy improves as it learns your venue's patterns

##### Override History

Review cases where team decisions differed from high-CCR recommendations

#### Venue Managers

Understand the score

##### See CCR Breakdown

Understand why Clementine gave a specific confidence rating to each recommendation

##### Contribute Feedback

Mark outcomes to help Clementine learn from your venue's specific results

##### Context Flags

Add context (local events, weather changes) that help refine future predictions

##### Review Low-CCR Tasks

Focus attention on items where Clementine needs human judgment

##### Accuracy Tracking

See your venue's prediction accuracy trends over time

#### Team Members

Act with confidence

##### See Confidence Indicators

Visual CCR badges on tasks show how certain OPPI is about recommendations

##### Prioritise by CCR

High-CCR tasks often indicate quick wins with reliable outcomes

##### Provide Outcome Data

Simple feedback on task results helps Clementine learn faster

##### Understand Reasoning

Clementine explains the factors behind its confidence ratings

##### Trust Building

See historical accuracy to build confidence in OPPI's suggestions