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%]
| Note |
|---|
The verbatim is analysed along with the question. It is therefore important to ask the right question to get a meaningful analysis. |
| Info |
|---|
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:
...
Number of positives
...
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
...
Nlp Personal Name
...
The value corresponds to a personal name. The value can contain the surname, first name or both.
...
Nlp Brand
...
The value corresponds to the name of a company in the public domain.
...
Nlp Product
...
The value corresponds to company products that are not necessarily in the public domain, but are detected as being products mentioned by the customer.
...
Nlp Location
...
The value corresponds to a physical location.
| Note |
|---|
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. |
| Info |
|---|
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 if you are manually adding themes to ensure effective 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.
Only administrators and editors can create/modify/delete a theme.
Choose topics that you’d like to track, if available, Feedier will group text responses that would fit under this topic type.
If you click on view the topic, you will only see the words of your topic appear in the ranking table, keyword cloud and text answers.
Topics will allow you to look for certain group of words and themes. For example, you can create a topic on Product Bugs and add words related to that topic.
...
If you want to look for a group of words, you can add the signs && in the search. To look for company excellence, you need to type in company&&excellence. The search will be showing every comment that contains the word company AND excellence.
...
You can always go back to add or remove words from the topic.
Similar articles on Text Analysis:
...
...
...
Summarisation of text responses
...
The Text Analysis module is comprised of three distinct components: the Responses per Topic component, the Sentiment Trend by Responses component, and a third component with three features: Summarisation by Eureka AI, Topic & Responses Tables, and Individual Topic Visualisation.
Responses per topic
The first component starting from the left is the Responses per topic component.This component consists in two different views of the responses of topics.
In the first tab we see a breakdown by the 10 topics with more volume considering the global Filters and the timeline dropdown (1)
How it works
...
Dropdown timeline to decide the time period to select the 10 topics to display.
This graph consists of a pie chart breakdown of the 10 topics
Legends of the graph
In the second tab we will see the same information but in an overtime graph
...
Dropdown timeline to decide the time period of the x-axis and to select the 10 topics to display.
This graph consists of a overtime chart of the 10 topics
Legends of the graph
...
Sentiment trend by responses
The second component from the left is the Sentiment Trend by Responses. This component features an overtime graph displaying the sentiment trend (positive, negative, and neutral) along with its volume.
How does it work
...
Dropdown timeline to decide the time period of the x-axis.
This graph consists of a overtime chart of the sentiment trend.
% of the total (based in global filter and dropdown timeline) of each sentiment score.
The topics with more volume related to the sentiment.
...
Tables & Summarisation
The next component we will discuss consists of three parts: an Eureka AI summary with insights of the text answers in your team, that also suggests five topics, a Topic Table offering valuable insights for each topic, and a Responses Table where you will be able to see each text answer and assign manually topics to each.
...
Eureka AI Summary
This component is the (1) of the previous image, after selecting the “Generate” button we will have the following.
...
Sentiment score of the text answers used to create the summary
Regenerate button to create a new summary
Eureka AI generated text
Suggested topics
Topics table
A table with a view of all the topics of the team.
...
Topics column → with the sentiment score badge
Sub topics column → with the sentiment score badge
The search bar to search for topics and sub topics
Responses column
Individual Visualisation View → we will see this after
3 dots
Edit topic
Delete topic
Amount of topics
Pagination
Individual Topic Visualisation
In this section, you will be able to explore various insights for each topic.
...
It consists of five different insights.
Breakdown of the topic by sentiment → % of responses that are positive/neutral/negative
Eureka AI Summary based on the topic
List of responses
Linked topics visualisation
List of sub topics
...
Responses Table
A table with a view of all the text answers of the team.
...
Text Answer list with the question and the survey
Sentiment score column of the text answers
Topics attached to the text answer
Add topic option to attach a topic manually to a text answer
Date of the text answer column
Link to open individual feedback of the text answer
...
Create a Topic
The create a topic flow consists in 3 steps:
The name of the topic → It’s important because this will be used for the attaching text answers to the topic by the AI
Parent topic selection in case you want to create a sub topic
Keywords → To expand to the name of the topic, other words you want to be related to a topic when present in a text answer
...
...