{
  "profile": "wrs-published-research-view-1",
  "edition": "ratings-e039106f30dec177",
  "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS",
  "entry": {
    "id": "cheng-et-al-wide-deep-recommendation-2016",
    "title": "Wide & Deep: joint sparse-feature and embedding-based recommendation",
    "summary": "Jointly trained a wide linear feature-cross model with a deep embedding network and evaluated the integrated ranking system in Google Play recommendations.",
    "boundary": "Inherits generalized linear models, feature crosses and embeddings. Memorization/generalization describe empirical modeling roles, not a theorem that their combination always improves recommendation. Online app-acquisition results are historical experiments, not present business impact or universal causal benefit; ranking is distinct from retrieval.",
    "area": "Recommendation and ranking systems",
    "topic": "Artificial intelligence",
    "year": 2016,
    "lane": "HUMAN",
    "result": "ALGORITHM",
    "package_id": "cheng-et-al-wide-deep-recommendation-2016",
    "target": {
      "record_id": "urn:uuid:05b25e18-beb9-46a9-a422-9a9a5c000352",
      "version": 1,
      "content_hash": "sha256:9fafb8c1b72b407af0957cb6a7ba597243dc14fbf39781add4d823f4cb082630"
    },
    "humans": [
      "Heng-Tze Cheng",
      "Levent Koc",
      "Jeremiah Harmsen",
      "Tal Shaked",
      "Tushar Chandra",
      "Hrishi Aradhye",
      "Glen Anderson",
      "Greg Corrado",
      "Wei Chai",
      "Mustafa Ispir",
      "Rohan Anil",
      "Zakaria Haque",
      "Lichan Hong",
      "Vihan Jain",
      "Xiaobing Liu",
      "Hemal Shah"
    ],
    "authors_complete": true,
    "human_role_note": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.",
    "agents": [],
    "related": [],
    "subresults": [],
    "scientific_assessment": {
      "source_review": "SOURCE_CURATED",
      "scientific_review": "NOT_REVIEWED_BY_WRS",
      "formal_replay": "NOT_RUN",
      "provider_attestation": "NONE"
    },
    "sources": [
      {
        "id": "cur20261015-arxiv-1606-07792v1",
        "title": "Wide & Deep Learning for Recommender Systems",
        "url": "https://arxiv.org/abs/1606.07792v1"
      }
    ],
    "points": "125000",
    "published_grade": {
      "id": "cheng-et-al-wide-deep-recommendation-2016",
      "entry_id": "cheng-et-al-wide-deep-recommendation-2016",
      "band": "A",
      "modifier": null,
      "target": {
        "record_id": "urn:uuid:05b25e18-beb9-46a9-a422-9a9a5c000352",
        "version": 1,
        "content_hash": "sha256:9fafb8c1b72b407af0957cb6a7ba597243dc14fbf39781add4d823f4cb082630"
      },
      "what_changed": "Jointly trained a wide linear feature-cross model with a deep embedding network and evaluated the integrated ranking system in Google Play recommendations.",
      "scope": "Jointly trained a wide linear feature-cross model with a deep embedding network and evaluated the integrated ranking system in Google Play recommendations. Inherits generalized linear models, feature crosses and embeddings. Memorization/generalization describe empirical modeling roles, not a theorem that their combination always improves recommendation. Online app-acquisition results are historical experiments, not present business impact or universal causal benefit; ranking is distinct from retrieval.",
      "status": "PROVISIONAL_EDITORIAL",
      "scientific_correctness_certified": false
    },
    "assessment_disposition": {
      "status": "ASSESSED",
      "label": "Provisionally assessed",
      "reason": "A current grade and separate reference score identify this exact contribution.",
      "target": {
        "record_id": "urn:uuid:05b25e18-beb9-46a9-a422-9a9a5c000352",
        "version": 1,
        "content_hash": "sha256:9fafb8c1b72b407af0957cb6a7ba597243dc14fbf39781add4d823f4cb082630"
      },
      "edition": "ratings-e039106f30dec177"
    }
  },
  "version": {
    "kind": "PUBLIC_RECORD_VIEW",
    "previous": null,
    "published_at": "2026-10-16T00:00:00Z",
    "publisher": {
      "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33",
      "role": "WRS_EDITORIAL_PUBLICATION"
    },
    "record": {
      "contributors": [
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Heng-Tze Cheng",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Levent Koc",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Jeremiah Harmsen",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Tal Shaked",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Tushar Chandra",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Hrishi Aradhye",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Glen Anderson",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Greg Corrado",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Wei Chai",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Mustafa Ispir",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Rohan Anil",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Zakaria Haque",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Lichan Hong",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Vihan Jain",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Xiaobing Liu",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        },
        {
          "attribution_basis": "SOURCE_CITED",
          "credit": "historical_scientific",
          "entity_kind": "historical_person",
          "name": "Hemal Shah",
          "note": "Archive attribution; detailed role assignment remains source-bound",
          "roles": [
            "formulator"
          ],
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        }
      ],
      "evidence": [
        {
          "description": "Bibliographic/stated-result source; not a WRS proof verification",
          "id": "source-0",
          "kind": "source_reference",
          "source": {
            "citation": "Wide & Deep Learning for Recommender Systems",
            "url": "https://arxiv.org/abs/1606.07792v1"
          }
        }
      ],
      "origin": "HUMAN_LEGACY",
      "record_profile": "wrs-local-research-notes-0.1",
      "record_type": "research_note",
      "rrp_version": "0.1",
      "scope": {
        "description": "Inherits generalized linear models, feature crosses and embeddings. Memorization/generalization describe empirical modeling roles, not a theorem that their combination always improves recommendation. Online app-acquisition results are historical experiments, not present business impact or universal causal benefit; ranking is distinct from retrieval.",
        "domain": "Recommendation and ranking systems"
      },
      "sources": [
        {
          "citation": "Wide & Deep Learning for Recommender Systems",
          "url": "https://arxiv.org/abs/1606.07792v1"
        }
      ],
      "statement": "# Wide & Deep: joint sparse-feature and embedding-based recommendation\n\nJointly trained a wide linear feature-cross model with a deep embedding network and evaluated the integrated ranking system in Google Play recommendations.\n\n## Scientific scope\n\nInherits generalized linear models, feature crosses and embeddings. Memorization/generalization describe empirical modeling roles, not a theorem that their combination always improves recommendation. Online app-acquisition results are historical experiments, not present business impact or universal causal benefit; ranking is distinct from retrieval.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nHeng-Tze Cheng; Levent Koc; Jeremiah Harmsen; Tal Shaked; Tushar Chandra; Hrishi Aradhye; Glen Anderson; Greg Corrado; Wei Chai; Mustafa Ispir; Rohan Anil; Zakaria Haque; Lichan Hong; Vihan Jain; Xiaobing Liu; Hemal Shah\n\nSource-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.\n\n\n\n## Original sources\n\n- [Wide & Deep Learning for Recommender Systems](https://arxiv.org/abs/1606.07792v1)\n",
      "summary": "Jointly trained a wide linear feature-cross model with a deep embedding network and evaluated the integrated ranking system in Google Play recommendations.",
      "title": "Wide & Deep: joint sparse-feature and embedding-based recommendation"
    },
    "record_id": "urn:uuid:05b25e18-beb9-46a9-a422-9a9a5c000352",
    "version": 1,
    "source_ref": {
      "record_id": "urn:uuid:05b25e18-beb9-46a9-a422-9a9a5c000352",
      "version": 1,
      "content_hash": "sha256:9fafb8c1b72b407af0957cb6a7ba597243dc14fbf39781add4d823f4cb082630"
    },
    "export_view": "wrs-public-editorial-view-1",
    "public_view_hash": "sha256:2457dda78fbd90679bdabc64cc6274921683d83dfd54ecc7635b5e7558002d9e"
  },
  "target": {
    "record_id": "urn:uuid:05b25e18-beb9-46a9-a422-9a9a5c000352",
    "version": 1,
    "content_hash": "sha256:9fafb8c1b72b407af0957cb6a7ba597243dc14fbf39781add4d823f4cb082630"
  },
  "published_grade": {
    "profile": "wrs-published-grade-1",
    "edition": "2026-10-28.1",
    "id": "cheng-et-al-wide-deep-recommendation-2016",
    "entry_id": "cheng-et-al-wide-deep-recommendation-2016",
    "revision": 1,
    "title": "Wide & Deep: joint sparse-feature and embedding-based recommendation",
    "context": "Recommendation and ranking systems",
    "track": "Human-designed research method or resource",
    "band": "A",
    "modifier": null,
    "status": "PROVISIONAL_EDITORIAL",
    "assessed_on": "2026-10-16",
    "evidence_cutoff": "2026-10-16",
    "achievement_state": "SUPPORTED_SCOPED_RESULT",
    "what_changed": "Jointly trained a wide linear feature-cross model with a deep embedding network and evaluated the integrated ranking system in Google Play recommendations.",
    "scope": "Jointly trained a wide linear feature-cross model with a deep embedding network and evaluated the integrated ranking system in Google Play recommendations. Inherits generalized linear models, feature crosses and embeddings. Memorization/generalization describe empirical modeling roles, not a theorem that their combination always improves recommendation. Online app-acquisition results are historical experiments, not present business impact or universal causal benefit; ranking is distinct from retrieval.",
    "evidence_summary": "The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Inherits generalized linear models, feature crosses and embeddings. Memorization/generalization describe empirical modeling roles, not a theorem that their combination always improves recommendation. Online app-acquisition results are historical experiments, not present business impact or universal causal benefit; ranking is distinct from retrieval.",
    "references": [
      {
        "title": "Wide & Deep Learning for Recommender Systems",
        "url": "https://arxiv.org/abs/1606.07792v1",
        "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification."
      }
    ],
    "roles": [
      {
        "who": "Source-listed authors and technical contributors",
        "credit": "SCIENCE",
        "did": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities."
      }
    ],
    "target": {
      "record_id": "urn:uuid:05b25e18-beb9-46a9-a422-9a9a5c000352",
      "version": 1,
      "content_hash": "sha256:9fafb8c1b72b407af0957cb6a7ba597243dc14fbf39781add4d823f4cb082630"
    },
    "scientific_correctness_certified": false,
    "independent_expert_calibration": false,
    "method": "PROPRIETARY",
    "points": "125000"
  },
  "canonical_agent_submission": false,
  "scientific_verification_awarded": false,
  "reader_guide": null,
  "reading_status": "SOURCE_LINKED_NOTE"
}