Abstract：Text mining based theme logic structure identificationapplication in library journals PurposeLibrary intelligence institutionswhich are a kind of traditional knowledge management organizationare at the frontline of the big data revolutionin which the use of unstructured data has become a modern knowledge management resourceThe paper aims to discuss this issueDesignmethodologyapproachThis research combined theme logic structureTLSartificial neural networkANNand ensemble empirical mode decompositionEEMDto transform unstructured data into a signal-wave to examine the research characteristicsFindingsResearch characteristics have a vital effect on knowledge management activities and management behavior through concentration and relaxationand ultimately form a quasi-periodic evolutionKnowledge management should actively control the evolution of the research characteristics because the natural development of six to nine years was found to be difficult to plotOriginalityvaluePeriodic evaluation using TLS-ANN-EEMD gives insights into journal evolution and allows journal managers and contributors to follow the intrinsic mode functions and predict the journal research characteristics tendencies Retrieval and IndexingandConsulting ServiceTheLibrary Intelligence Institutementioned in the article can be regarded as a specific knowledge management agencywhich is now challenging the big data revolutionThe use of non-structured data has become an important resource for modern knowledge managementThis study uses an artificial neural networkANNto convert non-structured data as signal waves to study its characteristicsANN is a technology of machine learningwhich allows data to learn and make predictions or decisions without artificial programming