Why does Feedier use artificial intelligence?
In the Feedier Platform, some of the data processed is unstructured data (textual feedback). Analysing this data can be tedious.
Today, Feedier uses artificial intelligence (AI) to carry out 3 types of analysis:
1: Sentiment analysis
2: Entity detection
3: Theme (topic) detection
General configuration
Feedier AI uses the context of the business to enable our models to deliver results that make sense in the operational context.
Go to your organisation page(Admin role)
Add an organisation name
Add a description of the organisation
It is important to complete the Description field on the "Organization" page. This description must be short and general. It can come directly from the website or the Wikipedia page.
Type 1 : Sentiment analysis
Each verbatim analysed is assigned to a sentiment: Positive, Neutral ou Negative.
Sentiment | Explanation | Sentiment rate |
|---|---|---|
| The verbatim is linked to a positive emotion on the part of the customer: Joy, Confidence, Serenity, Admiration. | [95%-100%] |
😐 | The verbatim is not directly linked to an emotion or the emotion is not expressed strongly enough to be categorised. | 50% |
😠 | The verbatim is linked to a negative emotion on the part of the customer: Anger, Contempt, Sadness, Disgust. | [1%-5%] |
The verbatim is analysed along with the question. It is therefore important to ask the right question to get a meaningful analysis.
Examples of bad questions: "Question Texte", "Q Texte", "Avis",
How is the sentiment rate calculated for a theme?
This is the average of the sentiment scores for each verbatim related to a theme.
Here are a few examples:
| 10 | 10 | 15 | 5 |
|---|---|---|---|---|
😐 Number of neutrals | 0 | 10 | 10 | 10 |
😠 Number of negatives | 0 | 10 | 5 | 10 |
Sentiment rate | ~ 100% | ~ 50% | ~ 66% | ~ 40% |
Type 2 : Entity detection
With Feedier, the "entities" extracted from the verbatims are added as attributes.
Attribute | Explanation |
|---|---|
| The value corresponds to a personal name. The value can contain the surname, first name or both. |
| The value corresponds to the name of a company in the public domain. |
| The value corresponds to company products that are not necessarily in the public domain, but are detected as being products mentioned by the customer. |
| The value corresponds to a physical location. |
The models may not detect 100% of entities and/or there may be a margin of error. Please do not hesitate to pass on any such information to the support@feedier.com team.
The labels (eg. Nlp Brand) displayed on the dashboard can be changed from the Attributes page.
Type 3 : Theme (topic) detection
Themes allow verbatims to be grouped according to common subjects. Once you have created a theme, you can:
Track the development of themes
Organise themes by sub-theme
Filter themes using the different Feedier filters
| Best practices for creating themes and ensuring the relevance of results: |
|---|---|
| We strongly recommend that you do not use keywords in addition to the AI to recognise themes. |
| We strongly recommend using short themes that encapsulate one idea at a time. |
Our commitments
We never use our customers' data to train the AI used (except in the case of a specific request in the "Feedier NLP" contract). The models are pre-trained upstream.
No customer data is saved outside the Feedier infrastructure when AI is used.
AI-related functionalities can be deactivated in Feedier, free of charge, on request to the Feedier support team(support@feedier.com).
0 Comments