Platform

Sentiment AI is an innovative platform designed for sentiment analysis in cryptocurrency communities. This product is a comprehensive tool intended for traders and investors who are interested in tracking and analyzing market sentiment in real time.

The platform's operation is based on collecting data from various social channels, such as Telegram and Twitter. This data is analyzed using special artificial intelligence algorithms that classify the emotional tone of messages and determine the overall sentiment in the cryptocurrency community.

The Sentiment AI platform provides users with the ability to quickly assess the market situation using data visualization in the form of diagrams and reports. This allows traders and investors to make informed decisions when trading and investing, minimizing risks and maximizing potential profits.

Sentiment AI can also be useful to analysts and researchers who are interested in market dynamics and are trying to understand the impact of community sentiments on the price dynamics of cryptocurrencies. The platform provides valuable information and analytics, helping users understand trends and factors affecting the price of cryptocurrencies in the near future.

Sentiment diagram

A key part of the platform is the sentiment diagram - a special data analysis visualization developed by Sentiment AI. This is a visualization of quantitative mood changes over time, shown in an interactive series of three interconnected diagrams.

How our Platform analyzes sentiment:

Sentiment probability

Every message written in the chat of Telegram, Twitter account or in Discord of our partners, is processed by our artificial intelligence model, which assigns it one of 7 sentiment indicators: 1) Fear 2) Skepticism 3) Greed 4) Neutrality 5) Happiness 6) Excitement 7) Trust.

Next, we select the timeframe and examine all the sentiment metrics derived from messages sent during this time period, and we find the probability of each metric appearing within this time period.

This probability distribution allows us to construct the first diagram showing the distribution.

Sentiment score

Next, we form the second graph with the distribution of emotions. For this, the model calculates the total number of messages, calculates the share of each of the 7 sentiments from the total number of messages, and creates a percentage distribution for specific sentiment. After this, weights are determined for each sentiment. These weights will be used to calculate the overall sentiment index.

The Sentiment score is calculated as a weighted sum of percentage shares of sentiment, multiplied by their weights and finally normalized in the range from 0 to 100. This process allows you to convert raw data about sentiment into a normalized score.

Sentiment index

After this, we visualize the influence of each sentiment on the Sentiment index - a message can decrease or increase it, resulting in a graph of sentiment changes over time.