convert simple question on different languages (english, german and french) to SQL queries using NLP

using NLP generate SQL queries for any database from simple search like what are the sales today compare iPhone sales with Samsung S10 everyday this year etc Detail of the project: It is about the analysis of a denormalized dataset by the SQL Select command That means there is no complex join. Since there is only one table where the data is denormalized. The complete synthax of a select must work. SELECT [ALL | DISTINCT] [AGREGAT FUNCTION], [Alias.] Select_Item [AS Column_Name] [, [Alias.] Select_Item [AS Column_Name] ...] FROM [DatabaseName!]Table [Local_Alias] [, DatabaseName!] Table [Local_Alias] ... [WHERE Filter Condition] [GROUP BY GROUPColumn [HAVING FilterCondition] [ORDER BY Order_Item [ASC | DESC] [, Order_Item [ASC | DESC] ...] The NLP logic should be so strong that e.g. the column name does not have to be explicitly present in the query in the natural language to identify the column automatically. For example, the following queries should be converted correctly to SQL. 1) What is the sales by product line ? 2) What is the evolution of sales by month? 3) how many sales did i get in EMEA by countries ? 4) what is the total of profit by type of product in usa? 5) what is the evolution of sales per city in the EMEA The code must be properly documented. I was programming in Java at the time. That means I favor Open NLP as NLP Lib. But if you think Python is better, e.g. NLTK, you can use the library. I have to be able to train my dataset in German, French and English without much programming. An interface via webservice is required after successful programming of the logic. In addition to the rule-based system, a learning system through machine learning must be implemented. Please contact me if you have any questions. We can discuss about the budget and project plan as well as milestones. Best regards Marcel