Description
Date depot: 15 juillet 2022
Titre: Blockchain token distribution strategies based on artificial intelligence
Directeur de thèse:
Farid NAIT-ABDESSELAM (LIPADE)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Systèmes et réseaux
Resumé: Blockchain technologies provide means to develop services that are secure, transparent, and efficient by nature. Unsurprisingly, the emerging business opportunities have gained a lot of interest that is realized in the form of successful token sales that are able to raise billions of USD. In the literature, there are two main approaches for token distribution, Initial Coin Offering (ICO),and bonding curve. In contrast to traditional exchanges, Automated Market Makers (AMMs) use bonding curves to provide automatic and continuous pricing models without the need to match buyers and sellers. However, this curve calculates the price depending solely on the supply. The aim of this work is thus to explore and utilize different attributes that can affect the price using Artificial Intelligence (AI) methods. Namely, we want to build a dynamic curve that takes into consideration the different factors that can affect the fluctuation of price. For example, but not limited to: Token Price in USD, Market price, Ethereum Price at launch, Date of launching the token sale, The quality of the development team, Popularity in social media, public opinion in twitter using sentiment analysis… etc. Furthermore, we will investigate the possibility of building an AI price prediction model above the bonding curve. This model will work like stock advisor, namely token sale advisor that will assist the investors in making wiser token investing decisions, is it worth the investment now? when to buy, and when to sell, and which organization to go with.
Résumé dans une autre langue: Blockchain technologies provide means to develop services that are secure, transparent, and efficient by nature. Unsurprisingly, the emerging business opportunities have gained a lot of interest that is realized in the form of successful token sales that are able to raise billions of USD. In the literature, there are two main approaches for token distribution, Initial Coin Offering (ICO),and bonding curve. In contrast to traditional exchanges, Automated Market Makers (AMMs) use bonding curves to provide automatic and continuous pricing models without the need to match buyers and sellers. However, this curve calculates the price depending solely on the supply. The aim of this work is thus to explore and utilize different attributes that can affect the price using Artificial Intelligence (AI) methods. Namely, we want to build a dynamic curve that takes into consideration the different factors that can affect the fluctuation of price. For example, but not limited to: Token Price in USD, Market price, Ethereum Price at launch, Date of launching the token sale, The quality of the development team, Popularity in social media, public opinion in twitter using sentiment analysis… etc. Furthermore, we will investigate the possibility of building an AI price prediction model above the bonding curve. This model will work like stock advisor, namely token sale advisor that will assist the investors in making wiser token investing decisions, is it worth the investment now? when to buy, and when to sell, and which organization to go with.
Doctorant.e: Safi Eljil Khouloud