Indicators for the Cryptocurrency Market
In the cryptocurrency market, investors often rely on various tools and methods to analyze market conditions in hopes of making wise investment decisions. Currently, common methods of market analysis include:
Technical Analysis: Technical analysis depends on historical price and trading volume data, using charts and technical indicators (such as moving averages, relative strength indices, Bollinger Bands, etc.) to predict future price trends. The fundamental assumption of technical analysis is that market prices reflect relevant information, thus investors use analysis of past price data to predict future price changes.
Fundamental Analysis: Fundamental analysis focuses on the intrinsic value of cryptocurrency projects, evaluating factors such as team background, technical strength, partnership relations, market demand, and competitive situations to assess the long-term potential of projects. Fundamental analysis is typically used to identify projects that have significant advantages in technology and market positioning for long-term investment.
On-chain Data Analysis: On-chain data analysis leverages the transparent nature of the blockchain to extract and analyze data from the blockchain network, such as transaction volume, wallet address distribution, hash rate, etc. This type of analysis provides profound insights into network health, user activity, and capital flow. In recent years, numerous outstanding on-chain data analysis products have emerged, offering a variety of on-chain analytical methods, such as nansen, coinglass, defillama, etc.
Sentiment Data Analysis: Sentiment data collection gathers market sentiments and public opinions from channels such as social media, news websites, and cryptocurrency forums. By monitoring these sentiments, investors can gauge the market's view and emotions towards specific projects. However, although there are many tools and websites for sentiment collection, their core capabilities often revolve around sentiment content collection, querying, and basic statistics, lacking technical means for emotional analysis and a systematic data indicator system, making it challenging for investors to extract useful market sentiment data from a vast amount of information.
What sources should sentiment data analysis include?
How can market and user attention as well as sentiment be standardized and quantified?
What kind of indicator system should be provided to users that can comprehensively and intuitively reflect the market and user evaluation of the project?
With these questions in mind, the Insight team developed the first version of Hermes, establishing a cryptocurrency sentiment analysis indicator system.
1 Sentiment Analysis Indicator
a) Overview of the Indicator System
The cryptocurrency sentiment analysis indicator system uses sentiment content from the cryptocurrency sector as raw data, which is then analyzed and quantified through a proprietary emotion analysis model to match scenario dimensions of market analysis, forming a comprehensive solution.
The system includes five core indicators:
Social Media Volume Index: An integrated evaluation of the project's engagement level on public social platforms, including metrics such as discussion intensity, attention level, emotional inclination, and key issues.
Community Volume Index: An integrated evaluation of the project's engagement level on private social platforms, including metrics such as discussion intensity, activity level, emotional inclination, and key topics.
Sentiment Index: An integrated evaluation of the emotional inclination of the project on core social media platforms.
Top Player Index: An integrated evaluation of the project's engagement by key institutions, top investors, and opinion leaders, including investment cooperation, holdings, discussions, and analysis.
Project Impact Index: An integrated evaluation of the project's influence, building on the foundational dimensions of fundamental analysis and incorporating market performance, media exposure, and community involvement to calculate the project's overall impact.
b) Data Sources
Insight, grounded in the science of social media, designs its data collection strategies starting from the core components of the sentiment ecosystem:
Information Sources: Ensuring coverage of mainstream sentiment sites within the cryptocurrency domain
Social media, including platforms like Twitter, Tiktok, YouTube, etc.
News media, encompassing major portals and news websites.
Forums and discussion groups, including Telegram, Discord, Reddit, etc.
Blogs and personal sites, including major research and analysis sites.
Official statements and announcements.
Discussion Content: Focusing on elements within the discussion content that provide key information
Events and Topics: The heart of sentiment discussions revolves around specific events or topics, such as news events, product launches, policy changes, etc.
Views and Opinions: Participants' perspectives and attitudes toward events or topics, including both positive endorsements and negative criticisms.
Emotions and Sentiments: Emotions expressed in discussions, such as anger, joy, worry, fear, etc., which significantly influence the direction and intensity of public sentiment.
Facts and Data: Facts, data, evidence, and cases cited in discussions, which help support or refute arguments.
Participants: Ensuring broad coverage of participants
General Public: Ordinary users are the primary participants in sentiment discussions, and their statements and interactions form the backbone of public sentiment.
Influencers: Influential individuals or groups, such as KOLs, whales, institutions, etc., whose opinions and actions often guide public sentiment.
Stakeholders: Individuals or organizations directly related to the event or topic, such as project teams, VCs, exchanges, etc., whose statements and actions also influence sentiment.
Communication Channels: Differentiating content acquisition based on communication channels
Social Media Platforms: Monitoring how content spreads through sharing, forwarding, commenting, etc., on social media.
News Media: Observing the spread and diffusion of news reports and commentaries.
Word-of-Mouth: Monitoring how sentiment information spreads among the public.
Automated Tools: Monitoring and removing the influence of recommendation algorithms on the objectivity of sentiment analysis, as bots and algorithmic recommendation systems play a significant role in sentiment dissemination, especially on social media platforms.
c) Design Principles
Insight aims to provide a truly objective, accurate, and practical indicator system, adhering to five key design principles: comprehensiveness, real-time, technical, standardization, and structure.
Comprehensiveness Principle: Provides a comprehensive portrait of market sentiment.
Data Sources: By gathering data from social media, news media, and forum discussions among other sources, it ensures the comprehensiveness and diversity of information.
Emotional Diversity: Not only focuses on the overall emotional tendency but also captures different types of emotions (such as positive, negative, neutral) and their intensity.
Layered Analysis: Differentiates discussions on community (public platforms) and community groups (private platforms), analyzing their popularity and emotions to provide more detailed insights.
Real-Time Principle: Ensures the timeliness of indicators through real-time data capture and dynamic updates.
Quick Market Response: Investors can obtain the latest market sentiment information in a timely manner, quickly adjusting investment strategies.
Continuous Monitoring: Ongoing sentiment monitoring helps to identify trends in emotional changes, avoiding decision-making errors due to information delays.
Technical Principle: Committed to embracing the latest technologies, continuously innovating and iterating the emotional analysis technology system.
Accuracy: The emotion analysis model can accurately distinguish different types of emotions and their intensity, providing more precise emotion analytics.
LLM Innovation: After the surge of LLM interest triggered by ChatGPT in the second half of 2022, LLM was promptly integrated to significantly upgrade the emotion analysis model with its superior capabilities in text and emotion analysis.
Standardization Principle: Ensures long-term consistency and accuracy of data indicators.
Panel Data Standardization: Indicators' longitudinal time series and cross-sectional data can undergo quantitative and trend analysis.
Algorithm Iteration Standardization: Any algorithm iteration ensures consistency with historical data.
Structuring Principle: Ensures the data indicator system has a strict and clear structure, unaffected by changes to the indicators.
Intuitiveness: Users can quickly grasp the overall situation and specific changes in market sentiment through intuitive indicators.
Detail Orientation: Each sub-indicator provides detailed emotional analysis dimensions, allowing users to deeply understand different aspects of market sentiment.
d) Working Principle
The Insight sentiment analysis indicator system starts from the underlying logic of sentiment propagation within social networks, including:
Sources and Origins: Sentiments are usually triggered by notable events or topics, such as breaking news, policy changes, celebrity statements, product launches, etc. Individuals or organizations with broad influence (such as celebrities, experts, media) play a significant role in the spread of sentiments. Their statements or actions often serve as the starting point or accelerant for public sentiment.
Information Propagation Paths: According to social network theory, the pathways of information propagation depend on the nodes (users) and edges (relationships) within social networks. Information spreads through the connections between nodes, forming complex network structures. Granovetter's "weak ties theory" highlights the important role of weak connections (such as acquaintances), as they link different social circles, enabling information to spread quickly across social groups.
Diffusion Mechanisms: The spread of information within social networks resembles viral propagation, relying on user actions such as sharing, forwarding, commenting, and interacting. Information that quickly attracts significant attention and interaction from many users will spread widely in a short period. The attractiveness, emotional resonance, practicality, and entertainment value of the content determine whether it can be widely disseminated. Information that evokes strong emotions and resonates with users is more likely to garner attention and spread.
Emotional Resonance and Amplification: Based on the theory of emotional contagion, emotions within social networks are easily influenced and transmitted among users. Positive or negative emotions can rapidly spread through interactions, creating emotional resonance. During the information propagation process, user feedback (such as likes, comments, sharing) further amplifies the impact of information, creating a positive feedback loop that quickly ferments public sentiment.
Credibility and Influence of Information: The effectiveness of information spread is influenced by the credibility of its source. Information from authoritative media, renowned experts, or credible organizations is more readily accepted and disseminated. The speed and extent of information spread are closely related to the network influence of the disseminator. High-influence users (such as KOLs) can significantly accelerate the spread of information.
Based on this, we can categorize the types of sentiment generation and propagation as follows:
Event-driven
Characteristics: Often triggered by sudden news such as official announcements or policy promotions.
Sentiment Indicator Trends: Rapid increase in discussion indicators, a one-sided emotional trend, with discussions focusing solely on the event itself without influencing the outcome. If there are no updates, the heat tends to decline.
Analysis Strategy: Focus on the direct impact and subsequent developments of the event, requiring rapid capture and response capabilities.
Role of Insight: By rapidly capturing breaking events through real-time monitoring and providing emotional analysis, it helps identify market sentiment shifts and assists investors in quickly adjusting their strategies to either hedge or seize short-term opportunities.
Celebrity-driven
Characteristics: Initiated by celebrities, these discussions carry significant momentum in information origin, propagation paths, and credibility, generating vibrant user discussions and fluctuating emotional indicators.
Sentiment Indicator Trends: Rapid rise in indicators that may remain high for an extended period. Emotional indicators fluctuate with active discussions and market sentiment changes, showing long-term persistence and potential for increase.
Analysis Strategy: Focus on the specific content and potential impact of celebrity statements, requiring real-time observation and monitoring within the topic's timeframe.
Role of Insight: Tracks celebrity statements and their sentiment impacts through real-time data capture, providing real-time emotional indicators to help investors identify fluctuations in market sentiment and make timely decisions and adjustments in celebrity-driven sentiments.
Operation-driven
Characteristics: Generally initiated by the project team or its partners, reflecting project developments and market strategies with strong planning and anticipation. This type of sentiment features a base increase in discussion indicators, with long-term persistence and growth potential, vigorous user discussions guided by the team, and a tendency in emotional indicators.
Sentiment Indicator Trends: Discussion indicators show a base lift with substantial long-term continuity and growth potential; user discussions are enthusiastic and guided by the team, leading to a directional development in emotional indicators.
Analysis Strategy: Focus on the positive extent and mid-to-long-term potential of the project news, as well as whether user opinions are positive and substantial in response to the news.
Role of Insight: Continuously monitors discussions and emotional changes through the sentiment analysis indicator system, providing long-term trend analysis to help investors identify the potential impacts of operational activities, discovering and positioning in potential projects early for long-term gains.
User-driven
Characteristics: Fermented by user communities, reflecting direct user experiences and evaluations of projects, characterized by high authenticity and interactivity but strong uncertainty. Discussion heat may quickly wane, and emotional indicators fluctuate significantly. This type of sentiment captures true user market conditions, initiated, fermented, and propagated by users, often carrying strong user consensus, broad profit potentials, and trends.
Sentiment Indicator Trends: Gradual increase in sentiment indicators, with risks of rapid quieting; emotional indicators fluctuate significantly, with heat trends influenced by user participation and interaction instability.
Analysis Strategy: Capable of promptly identifying sentiments that reach a certain level of discussion and continuously monitoring whether subsequent discussions can maintain an upward trend and whether there are significant changes in user emotions.
Role of Insight: While this type of sentiment has a massive base and low probability of generating significant market conditions, traditional analysis tools require considerable time. With the Insight sentiment analysis indicator system, it can quickly identify the development stages of each sentiment through indicators and trends, filtering the market conditions more efficiently.
2 Additional Indicators
While Insight's emotional analysis indicator system provides crucial reference data in market analysis, it is only a core component and not a panacea. Emotional analysis data must be integrated with other indicators and analytical methods to derive truly effective and accurate conclusions. Therefore, Insight's cryptocurrency market indicator system has also developed multiple sets of other analytical methodologies, forming a unique matrix of indicator systems. These systems work together to provide multi-dimensional, multi-layered market insights, assisting users in making more scientific and comprehensive investment decisions. These include:
On-chain Analysis Indicator System: Utilizes the transparent nature of blockchain technology to extract and analyze data from the blockchain network, offering profound insights into network health, user activity, and fund flows. This includes transaction data, wallet address data, miner data, and smart contract data.
Cryptocurrency Project Market Analysis Indicator System: Provides in-depth project analysis and comparison tools for investors by assessing the market performance and potential of cryptocurrency projects. This includes basic project information, market performance data, financing data, development activity data, and community activity data.
Macroeconomic and Sector Economic Indicator System: Offers comprehensive market background analysis and trend forecasting for investors by analyzing macroeconomic environments and specific sector economic data. This includes macroeconomic data, industry data, policy and regulation data, and global market data.
Last updated