From a diamond registry to passport control systems, blockchain technology is already being applied to a wide array of human activities by various companies. One of these firms is Hong Kong-based Senno.
Founded in January of 2017, sometime before the cryptocurrency boom really took off, Senno is seeking to use blockchain technology to perform real-time sentiment analysis. Based on the NEO blockchain, a Chinese blockchain system, the firm will use its own tokens to reward participants who support the development of its service.
In the long run, the data Senno produces could be used in a wide variety of industries and business functions, whether it be marketing for a food company or risk management for a bank.
For now, the firm is focusing its efforts on the financial markets. In light of this, Finance Magnates spoke to Elad Peled this Thursday, Senno’s co-founder, and CEO, about listing Senno’s tokens on exchanges, product development and fears surrounding personal data.
FM: When one thinks of ‘blockchain’ or ‘cryptocurrency,’ the first thing that comes to mind is certainly not sentiment analysis, and yet that is what Senno is looking to provide. What sparked your interest in sentiment analysis and how did you come up with the idea of mapping it on to a blockchain-based system?
EP: About ten years ago, I was involved in a project that used sentiment analysis to predict stocks trends in traditional markets, but the accuracy level wasn’t satisfying, and the hardware used for calculation was too expensive. When I first started to trade cryptocurrencies, I noticed that public speculations and opinions play significant roles in market trends and using sentiment analysis to process this data will have a much greater effect than in traditional markets. In addition, blockchain technology allows us to distribute hardware resources and create a contributors economy that can significantly reduce hardware costs.
FM: You have plans to launch Senno tokens on to the major crypto-exchanges. How are those efforts going – can we expect to see Senno’s tokens being traded on some of major exchanges soon?
EP: Correct, we are currently in discussions with several major exchanges, although we will not be able to disclose any names until finalized, we estimate that the listing will take place before the end of August.
FM: A major part of your service offering seems to be reduced costs that are a result of Senno members contributing their own computer power to Senno’s blockchain network, as opposed to the company having to maintain a large data center. Are there any risks to this? For instance, you plan to pay these people in Senno tokens but what if users feel that the value of the token doesn’t match the power they are generating for you? What if you simply can’t muster enough users to generate the required power or the number of users fluctuates to a point that your service becomes unreliable?
EP: Yes, there are always risks in maintaining a complete ecosystem; we have taken several measurements in order to make sure there will always be a financial incentive which will maintain the completeness of the network; the payment for Senno’s data will be calculated according to queries complexity, thus inherently reflecting the costs needed to perform the calculations and incentivize contributors. In addition, 30% of Senno’s token supply will be kept by the smart contracts in charge of rewarding and will be used to further incentivize contributes in the early days of the network.
FM: Senno also plans on using its tokens to pay for access to their data. That data would then be used to gauge public sentiment. Some of this data, from instant messaging services, for example, is likely to be very personal. How can you convince people that it is worth handing over their data to you and that their data will be safe?
FXTM Recruits Financial Broadcaster Han Tan to its Market Research TeamGo to article >>
EP: With all the recent scandals surrounding the social web, we believe users should have the opportunity to share their private data and be compensated for it. The data is physically stored in a decentralized storage space, protected by ECC asymmetric encryption, sharding techniques, and various anti-sybil attack methods. This guarantees that a Cambridge Analytica-like incident can’t happen on Senno, while users can enjoy complete data ownership.
FM: Public attitudes towards data privacy vary dramatically across the globe. Europeans, for example, are very concerned about data privacy whereas the Chinese have a much more ‘laissez-faire’ attitude towards it. Do you worry that the data you gather could be skewed towards countries that have a more lax attitude towards data privacy and this, in turn, could affect your sentiment analysis efforts?
EP: Using Artificial Neural Network (ANN), Senno’s Machine learning algorithms will be able to segregate high value data sources identified by the NEO digital identity protocols and grant them with higher internal scoring, in a case where certain data sources will skew the overall analysis to a false conclusion their scoring will be adjusted accordingly and have lesser relative effect on the sentiment result.
FM: Will you make efforts to segment data according to the needs of the person accessing it? For instance, within the trading industry alone, someone trading in Japanese equities is likely to need a very different set of data to someone trading in the USD/TRY FX market.
EP: Yes, The Senno SDK and API allow third parties to connect to the network and execute dynamic data queries to receive tailored data according to their subject of interest.
FM: Is there any danger inherent to large-scale sentiment analysis. In the trading world, for example, a herd mentality can lead to market bubbles, could a service such as yours accentuate that?
EP: A large-scale sentiment analysis will inherently lead to earlier identification of market trends, thus accelerating the process of both creation and deterioration of potential market bubbles. Basically, the use of such technology will cause the market to respond quicker and faster than ever before.
FM: Senno describes itself as providing ‘sentiment analysis.’ There seems to be a big difference, however, between gauging public sentiment and analyzing it. For example, it’s interesting to know that a million people are discussing bitcoin but how would Senno translate that information into meaningful data for traders?
EP: Senno answers this question by generating two different indexes from the analysed data – Buzz & Mood, the Buzz index will demonstrate the amount of discussion made on a certain topic (e.g 1M discussions on bitcoin in the last 24 hour period) while the Mood index will demonstrate the overall sentiment value of those discussions (e.g. 85% positive). Senno will also present the delta changes on those indexes (delta Buzz and delta Mood) compare with previous time period. These capabilities are already available for review on Senno’s MVP which was just released – https://mvp.senno.io/.