ALOC – At Least One Connection (Semeion©)
The A.L.O.C. system represents a new and semantically relevant mode for finding hidden relationships between many datasets, different in terms of variables and records, but linked to the same reality.
The A.L.O.C. system operates in accordance with the principle common to all the semantic memories belonging to the biological type: given a detail the system reconstructs its context (other variables), and is able to list the typical experiences (records) that support the validity of that reconstruction.
The A.L.O.C. system dynamically creates links between variables from different datasets according to the stimuli that it receives. These links are modified during the answering process according to the fuzzy similarity between variables and records, and to the competition and cooperation between the variables themselves. This process also allows the A.L.O.C. system to propose combinations of variables that are not present in any record of any datasets processed in that mode. However, these original combinations describe the best prototype that can be generated from the initial links (“external input”). Therefore, the A.L.O.C. system is able to make basic abstractions from the data.
 P.M. Buscema
Artificial Adaptive System for Parallel Querying of Multiple Databases
chapter 19, pp 481-511, in M. Buscema and W Tastle (eds.), Intelligent Data Mining in Law Enforcement Analytics:New Neural Networks Applied to Real Problems, DOI 10.1007/978-94-007-4914-6 19, © Springer Science+Business Media Dordrecht 2013.