{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:ISMDEYY5I2JL35B6TNNTKD7J5L","short_pith_number":"pith:ISMDEYY5","schema_version":"1.0","canonical_sha256":"449832631d4692bdf43e9b5b350fe9ead882ffaa1c5a185a1a6d158a7aaed4ee","source":{"kind":"arxiv","id":"1809.04019","version":1},"attestation_state":"computed","paper":{"title":"Training and Prediction Data Discrepancies: Challenges of Text Classification with Noisy, Historical Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","stat.ML"],"primary_cat":"cs.IR","authors_text":"Emilia Apostolova, R. Andrew Kreek","submitted_at":"2018-09-11T16:43:52Z","abstract_excerpt":"Industry datasets used for text classification are rarely created for that purpose. In most cases, the data and target predictions are a by-product of accumulated historical data, typically fraught with noise, present in both the text-based document, as well as in the targeted labels. In this work, we address the question of how well performance metrics computed on noisy, historical data reflect the performance on the intended future machine learning model input. The results demonstrate the utility of dirty training datasets used to build prediction models for cleaner (and different) predictio"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1809.04019","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2018-09-11T16:43:52Z","cross_cats_sorted":["cs.CL","cs.LG","stat.ML"],"title_canon_sha256":"6b1357d419141c6d1995ae2819305d9dbab86c1ca8dca5be35997106936edb75","abstract_canon_sha256":"3578b97b48c8d645313b40b14d24abd4b0e1dfd2a3bb6195d14b6cc78577d510"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:05:57.515473Z","signature_b64":"JRJtVnZV4GskCuvmgtZbDf/YQY37KXBKCxc+/g2F6zkpVmkOiJXtiBna0KaO08qbihCkD1qrOrFGLUhwcG1LDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"449832631d4692bdf43e9b5b350fe9ead882ffaa1c5a185a1a6d158a7aaed4ee","last_reissued_at":"2026-05-18T00:05:57.514984Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:05:57.514984Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Training and Prediction Data Discrepancies: Challenges of Text Classification with Noisy, Historical Data","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","stat.ML"],"primary_cat":"cs.IR","authors_text":"Emilia Apostolova, R. Andrew Kreek","submitted_at":"2018-09-11T16:43:52Z","abstract_excerpt":"Industry datasets used for text classification are rarely created for that purpose. In most cases, the data and target predictions are a by-product of accumulated historical data, typically fraught with noise, present in both the text-based document, as well as in the targeted labels. In this work, we address the question of how well performance metrics computed on noisy, historical data reflect the performance on the intended future machine learning model input. The results demonstrate the utility of dirty training datasets used to build prediction models for cleaner (and different) predictio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1809.04019","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"1809.04019","created_at":"2026-05-18T00:05:57.515061+00:00"},{"alias_kind":"arxiv_version","alias_value":"1809.04019v1","created_at":"2026-05-18T00:05:57.515061+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1809.04019","created_at":"2026-05-18T00:05:57.515061+00:00"},{"alias_kind":"pith_short_12","alias_value":"ISMDEYY5I2JL","created_at":"2026-05-18T12:32:31.084164+00:00"},{"alias_kind":"pith_short_16","alias_value":"ISMDEYY5I2JL35B6","created_at":"2026-05-18T12:32:31.084164+00:00"},{"alias_kind":"pith_short_8","alias_value":"ISMDEYY5","created_at":"2026-05-18T12:32:31.084164+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.15547","citing_title":"Consistency Analysis of Sentiment Predictions using Syntactic & Semantic Context Assessment Summarization (SSAS)","ref_index":23,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ISMDEYY5I2JL35B6TNNTKD7J5L","json":"https://pith.science/pith/ISMDEYY5I2JL35B6TNNTKD7J5L.json","graph_json":"https://pith.science/api/pith-number/ISMDEYY5I2JL35B6TNNTKD7J5L/graph.json","events_json":"https://pith.science/api/pith-number/ISMDEYY5I2JL35B6TNNTKD7J5L/events.json","paper":"https://pith.science/paper/ISMDEYY5"},"agent_actions":{"view_html":"https://pith.science/pith/ISMDEYY5I2JL35B6TNNTKD7J5L","download_json":"https://pith.science/pith/ISMDEYY5I2JL35B6TNNTKD7J5L.json","view_paper":"https://pith.science/paper/ISMDEYY5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1809.04019&json=true","fetch_graph":"https://pith.science/api/pith-number/ISMDEYY5I2JL35B6TNNTKD7J5L/graph.json","fetch_events":"https://pith.science/api/pith-number/ISMDEYY5I2JL35B6TNNTKD7J5L/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ISMDEYY5I2JL35B6TNNTKD7J5L/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ISMDEYY5I2JL35B6TNNTKD7J5L/action/storage_attestation","attest_author":"https://pith.science/pith/ISMDEYY5I2JL35B6TNNTKD7J5L/action/author_attestation","sign_citation":"https://pith.science/pith/ISMDEYY5I2JL35B6TNNTKD7J5L/action/citation_signature","submit_replication":"https://pith.science/pith/ISMDEYY5I2JL35B6TNNTKD7J5L/action/replication_record"}},"created_at":"2026-05-18T00:05:57.515061+00:00","updated_at":"2026-05-18T00:05:57.515061+00:00"}