MIKA

The MIKA package contains three core components for analyzing large sets of narrative text: mika.kd for knowldege discovery, mika.ir for information retrieval, and mika.utils for utilities. Typically, a user will load and configure their data using mika.utils and then use a combination of functions from mika.kd and mika.ir to perform an analysis.

• utils package
  • mika.utils.Data
    • Data
      • Data.load()
      • Data.prepare_data()
      • Data.preprocess_data()
      • Data.save()
      • Data.sentence_tokenization()
• kd package
  • mika.kd.FMEA
    • FMEA
      • FMEA.build_fmea()
      • FMEA.calc_frequency()
      • FMEA.calc_risk()
      • FMEA.calc_severity()
      • FMEA.display_doc()
      • FMEA.evaluate_preds()
      • FMEA.get_entities_per_doc()
      • FMEA.get_year_per_doc()
      • FMEA.group_docs_manual()
      • FMEA.group_docs_with_meta()
      • FMEA.load_data()
      • FMEA.load_model()
      • FMEA.post_process_fmea()
      • FMEA.predict()
  • mika.kd.Topic_Model_plus
    • Topic_Model_plus
      • Topic_Model_plus.bert_topic()
      • Topic_Model_plus.calc_bert_coherence()
      • Topic_Model_plus.coherence_scores()
      • Topic_Model_plus.get_bert_coherence()
      • Topic_Model_plus.get_bert_topic_diversity()
      • Topic_Model_plus.get_bert_topics_from_probs()
      • Topic_Model_plus.hdp()
      • Topic_Model_plus.hlda()
      • Topic_Model_plus.hlda_display()
      • Topic_Model_plus.hlda_extract_models()
      • Topic_Model_plus.hlda_visual()
      • Topic_Model_plus.label_hlda_topics()
      • Topic_Model_plus.label_lda_topics()
      • Topic_Model_plus.lda()
      • Topic_Model_plus.lda_extract_models()
      • Topic_Model_plus.lda_visual()
      • Topic_Model_plus.load_bert_model()
      • Topic_Model_plus.reduce_bert_topics()
      • Topic_Model_plus.save_bert_coherence()
      • Topic_Model_plus.save_bert_document_topic_distribution()
      • Topic_Model_plus.save_bert_model()
      • Topic_Model_plus.save_bert_results()
      • Topic_Model_plus.save_bert_taxonomy()
      • Topic_Model_plus.save_bert_topic_diversity()
      • Topic_Model_plus.save_bert_topics()
      • Topic_Model_plus.save_bert_topics_from_probs()
      • Topic_Model_plus.save_bert_vis()
      • Topic_Model_plus.save_hlda_coherence()
      • Topic_Model_plus.save_hlda_document_topic_distribution()
      • Topic_Model_plus.save_hlda_level_n_taxonomy()
      • Topic_Model_plus.save_hlda_models()
      • Topic_Model_plus.save_hlda_results()
      • Topic_Model_plus.save_hlda_taxonomy()
      • Topic_Model_plus.save_hlda_topics()
      • Topic_Model_plus.save_lda_coherence()
      • Topic_Model_plus.save_lda_document_topic_distribution()
      • Topic_Model_plus.save_lda_models()
      • Topic_Model_plus.save_lda_results()
      • Topic_Model_plus.save_lda_taxonomy()
      • Topic_Model_plus.save_lda_topics()
      • Topic_Model_plus.save_mixed_taxonomy()
  • mika.kd.NER
    • align_labels_with_tokens()
    • build_confusion_matrix()
    • check_doc_to_sentence_split()
    • clean_annots_from_str()
    • clean_doccano_annots()
    • clean_text_tags()
    • compute_classification_report()
    • compute_metrics()
    • get_cleaned_label()
    • identify_bad_annotations()
    • plot_eval_metrics()
    • plot_eval_results()
    • plot_loss()
    • read_doccano_annots()
    • read_trainer_logs()
    • split_docs_to_sentences()
    • tokenize()
    • tokenize_and_align_labels()
  • mika.kd.trend_analysis
    • add_hazards_to_docs()
    • bootstrap_metric()
    • build_word_clouds()
    • calc_CI()
    • calc_classification_metrics()
    • calc_rate()
    • calc_severity_per_hazard()
    • check_for_hazard_words()
    • check_for_negation_words()
    • chi_squared_tests()
    • corr_sig()
    • create_correlation_matrix()
    • examine_hazard_extraction_mismatches()
    • get_doc_text()
    • get_doc_time()
    • get_hazard_df()
    • get_hazard_doc_ids()
    • get_hazard_info()
    • get_hazard_topics()
    • get_hazard_topics_per_doc()
    • get_hazard_words()
    • get_likelihood_FAA()
    • get_likelihood_USFS()
    • get_negation_words()
    • get_results_info()
    • get_topics_per_doc()
    • get_word_frequencies()
    • hazard_accuracy()
    • identify_docs_per_fmea_row()
    • identify_docs_per_hazard()
    • make_pie_chart()
    • minmax_scale()
    • multiple_reg_feature_importance()
    • plot_USFS_risk_matrix()
    • plot_frequency_time_series()
    • plot_metric_averages()
    • plot_metric_time_series()
    • plot_predictors()
    • plot_risk_matrix()
    • proposed_topics()
    • record_hazard_doc_info()
    • regression_feature_importance()
    • remove_outliers()
    • reshape_correlation_matrix()
    • sample_for_accuracy()
    • set_up_docs_per_hazard_vars()
• ir package
  • mika.ir.search
    • search
      • search.get_sentence_embeddings()
      • search.load_sentence_embeddings()
      • search.run_search()