We aggregate information from public online communities across instant messaging, microblogging and social network platforms. Hundreds of millions of records are retrieved and run through several GDPR-compliant data cleaning, anonymisation and verification algorithms to ensure data integrity before being organised and analysed.
We harness this massive amount of unstructured data and provide accurate, timely and unbiased analytics of social media content using proprietary algorithms and methodologies in the area of data mining and artificial intelligence including machine learning neural networks and deep learning.
The analysis is synthetised in proprietary indexes capturing unique social media dynamics.
These tools currently include datasets with daily data from platforms such as Telegram and Reddit, but an upcoming version will include also Twitter, Stocktwits, Yahoo Finance, Github, Coinmarketcap chat and BitcoinTalk.
The Trust Index is an unbiased synthetic indicator which summarises the overall fairness and quality of chat group and forum discussions. The Trust Index is obtained aggregating several other sub-indexes calculated by taking into the equation the statistical and network properties of the following quantitative measures: Users activity; Messages; Bots; Sensitive words; association to other low quality chat groups; chat administrator and moderator performance. Some, but not all, of the main quantitative dimensions represented can be connected to the engagement of the chat group participants, the responsiveness of the administrators, the reliability of the users involved, the quality of the content generated and the soundness of the idiom used
The Mood Index provides sentiment-driven cryptocurrency signals. We built a proprietary sentiment dictionary to estimate sentiment across tens of thousands of open chats and forum in the crypto space. We identify emotional properties embodied in the messages together with a measure of the confidence in our estimates.
The Scam Index is a proprietary unbiased measure of the level of scams and swindles in online social media communities. We use proprietary metrics and Machine Learning algorithms to develop online fraud forensics and detect phishing scams, impersonations - fake accounts, worms, bot spams, money-based schemes and Ponzi schemes.
The Spam Index measures the level of spamming activities in online social media forums and chat groups. We use proprietary metrics and Deep Learning algorithms and Computer Visions techniques to detect irrelevant or unsolicited messages in online forums and chats group discussions.
Emerging digital technologies like DLT have the potential to drive positive change in a wide range of Environmental, Social, and Governance (ESG) issues. However, the energy consumption required by some blockchain networks has raised concerns about their carbon footprint and impact on the environment. As such, tracking and ranking blockchains based on their carbon footprint can help to incentivise more sustainable practices and promote the adoption of greener alternatives, making DLT an even more powerful force for positive change in the future.
In the spirit of the DSFs commitment to maximising DLT’s positive impact, to share knowledge and to foster the open and transparent development of digital ledger ecosystems we have developed proprietary tools measuring:
On-chain analytics tools
The DLT Science Foundation is developing advanced on-chain data analytics tools that will provide invaluable real-world insights into DLTs. This powerful tool will offer deep insights into ledger phenomena, including:
This will provide perspectives on network health and profitability, market trends, user sentiment, and risk management. Metrics such as validator performance and rewards distribution will enable stakeholders to better understand network status and even inform resource allocation decisions. Additionally, the tools will give users a detailed comprehension of the flow of value within blockchain networks, helping them identify key players and trends.