cic-docs/TransactionDatasets.md

# TransactionDatasets
Data Sets from Blockchain Transactions and User Meta Data found at 
[grassecon.org/research](http://grassecon.org/research) in csv format and also shown on [dashboard.sarafu.network](https://dashboard.sarafu.network)

Datasets prior to 2020 are pulled from the POA blockchain for each user ex. https://blockscout.com/poa/core/address/0xcb55fc893000a984a5ad73011f93d1540a5f0895/tokens
As well as user generated data and field surveys collected in Kenya. Fields described below.

Datasets Starting 2020 are pulled from the xDAI blockchain for each user ex. https://blockscout.com/poa/core/address/0xcb55fc893000a984a5ad73011f93d1540a5f0895/tokens
As well as user generated data and field surveys collected in Kenya. Fields described below.

Brief description: Using a simple feature phone interface with USSD each user (people living below the poverty line in Kenya) starts with an allottment of initial tokens and adverstises what they sell to neighbors and on a digital market place avalaible on the same interface (USSD). There are no transaction fees. The tokens are used often as a top-up to missing Kenyan shillings. Pre 2020 - each user only has one token assigned to them by GE based on where they live. Post 2020 all users moved to the Sarafu token while we prepare for user generated tokens again. Pre 2020 with multiple tokens - When a user sends a token to another user that holds a different token a conversion is made automatically after receipt. This conversion uses a bonding curve (based on the Bancor Protocol) to adjust the token prices relative to eachother. Conversions can be seen as transactions to a contract address then that contract address sending the new token to the user. In 2020 the Sarafu token which all users held is considered ~1:1 with the national currency. For more information please contact us https://www.grassrootseconomics.org/contact


--------

Post 2020 xDAI data (current)
The Transaction csv fields:

1. id - internal transaction ID number
1. timeset - date and time of transaction
1. transfer_subtype - internal typing: DISBURSMENT = from Grassroots Economics Foundation, RECLAMATION = Back to GE, STANDARD =  a trade between users, AGENT = when a group account is cashing out (see held_roles) below
1. source - wallet ID of sender 
1. target - wallet ID of recipent 
1. token_name - the token that was traded
1. token_address - the blockchain address of token that was traded
1. weight - How many tokens were traded

The user summary csv fields:

1. id - Internal user id number
1. start - day and time of first transaction (when the user account was setup)
1. xDAI_blockchain_address - Wallet ID on POA xDAI Blockchain 
1. old_POA__blockchain_address - Wallet ID on the older POA Blockchain (pre 2020) if they had one
1. final_bal - current balance of Sarafu as of file date
1. area_name - Regional area in Kenya (generalized from private user data)
1. area_type - Rural, Peri urban or Urban (generalized from private user data)
1. held_roles - Standard transactions are between Beneficiaries, anything to an Admin is a Reclamation or from and Admin is a Disbursment and anything to a Agent is a Agent_out
1. gender - user input gender
1. business_type - Input by GE staff based on what the users sell
1. ovol_in - total number of tokens that came into this account from non-STANDARD transactions
1. ovol_out - total number of tokens that left this account from non-STANDARD transactions
1. otxns_in - number of transactions incomming from non-STANDARD transactions
1. otxns_out
1. ounique_in - number of uniquie transactions incomming from non-STANDARD transactions
1. ounique_out
1. svol_in - total number of tokens that came into this account from STANDARD transactions
1. svol_out
1. stxns_in - number of transactions incomming from STANDARD transactions
1. stxns_out
1. sunique_in - number of uniquie transactions incomming from STANDARD transactions
1. sunique_out


----

Pre 2020 POA
The Transaction csv fields:
1. timeset - date and time of transaction
1. tx_hash - hashed transaction on blockchain can be found on https://blockscout.com/poa/core/a
1. source - wallet ID of sender can be found on https://blockscout.com/poa/core/a - which could be a contract (for a conversion)
1. s_comm_tkn - Short name of Token that the source uses (each user should only have 1 token)
1. s_gender - From user input
1. s_location - From user input
1. s_business_type - Grassroots Economics Staff
1. s_directory - From user input
1. target - wallet ID of recipent - which could be a contract (for a conversion)
1. t_comm_tkn
1. t_gender
1. t_location
1. t_business_type
1. t_directory
1. tx_token - the token that was traded
1. weight - How many tokens were traded
1. id - test variable


The user summary csv fields:
1. id - Wallet ID on POA Blockchain - can be found on https://blockscout.com/poa/core/a
1. comm_tkn - Their token name (each user only has 1 token)
1. bal - current balance of [comm_tkn] as of file date
1. location - User input (village name)
1. gender
1. business_type - Input by staff based on what they sell
1. directory - input by user a specific item they sell
1. vol_trans_in - total number of tokens that came into this account
1. vol_trans_out
1. n_trans_in - number of transactions incomming
1. n_trans_out
1. n_in_unique - number of uniquie transactions incomming
1. n_out_unique
1. start - day and time of first transaction (when the user account was setup)