Method and apparatus for trade interaction chain reconstruction

A method for trade interaction chain reconstruction, comprising:

• receiving audio interactions and text interactions by a computerized system executed on one or more general purpose computers;
• pre-processing the interactions by extracting words from the audio interactions and metadata from the audio interactions and the text interactions;
• wherein the metadata includes an order ID, an execution ID or a swap deal ID;
• storing the pre-processed interactions and results of the pre-processing in a storage device;
• querying the storage device with an expression to identify a swap deal, the swap deal includes two or more of the received interactions and involves two or more participants;
• assigning a relevance score of each interaction to the query expression;
• sorting the interactions by the relevance score;
• automatically selecting from the storage device based on the relevance score a first interaction of the received interactions, said first interaction involves at least two participants of the two or more participants;
• obtaining a first plurality of interactions of the received interactions that involve the at least two participants of the two or more participants;
• filtering the obtained first plurality of interactions based on metadata parameters;
• determining a first plurality of relevance scores between the first plurality of interactions and the first interaction;
• and associating interactions of the first plurality of interactions to be relevant to the swap deal according to the determined first plurality of relevance scores
• outputting to a chain storage module a potential list of interactions that were determined to be relevant to the swap deal;
• wherein the preprocessing comprises:
• automatically transcribing the received audio interactions;
• extracting words, phrases and named entities from the received interactions;
• associating metadata to the receive interactions;
• categorizing the received interaction into categories;
• wherein the named entities include names of people, names of organizations or names of locations;
• and
• wherein the categories are identified by assigning a category score that is a function of a log of one plus the inverse of the sum of the number of extracted words, extracted named entities and extracted metadata or a specific interaction.