MétaCan
Menu
Back to cohort
Record W3047519683 · doi:10.48336/kbz2-sv47

Attribute lattice: a graph-based conceptual modeling grammar for heterogeneous data

2021· dissertation· en· W3047519683 on OpenAlexaff
Mojtaba Asgari

Bibliographic record

VenueMemorial University Research Repository (Memorial University) · 2021
Typedissertation
Languageen
FieldComputer Science
TopicSemantic Web and Ontologies
Canadian institutionsMemorial University of Newfoundland
Fundersnot available
KeywordsComputer scienceNatural language processingTheoretical computer scienceArtificial intelligenceInformation retrieval

Abstract

fetched live from OpenAlex

One key characteristic of big data is variety. With massive and growing amounts of data existing in independent and heterogeneous (structured and unstructured) sources, assigning consistent and interoperable data semantics, which is essential for meaningful use of data, is an increasingly important challenge. I argue, conceptual models, in contrast to their traditional roles in the Information System development, can be used to represent data semantics as perceived by the user of data. In this thesis, I use principles from philosophical ontology, human cognition (i.e., classification theory), and graph theory to offer a theory-based conceptual modeling grammar for this purpose. This grammar reflects data from users of data perspective and independent from data source schema. I formally define the concept of attribute lattice as a graph-based, schema-free conceptual modeling grammar that represents attributes of instances in the domain of interest and precedence relations among them. Each node in an attribute lattice represents an attribute - a true statement (predicate) about some instances in the domain. Each directed arc represents a precedence relation indicating that possessing one attribute implies possessing another attribute. In this thesis, based on the premise that inherent classification is a barrier that hinders semantic interoperation of heterogeneous data sources, a human cognition based conceptual modeling grammar is introduced as an effective way to resolve semantic heterogeneity. This grammar represents the precedence relationship among attributes as perceived by human user and provides a mechanism to infer classes based on the pattern of precedences. Hence, a key contribution of attribute lattice is semantic relativism – that is, the classification in this grammar relies on the pattern of precedence relationship among attributes rather than fixed classes. This modeling grammar uses the immediate and semantic neighbourhoods of an attribute to designate an attribute as either a category, a class or a property and to specify the expansion of an attribute – attributes which are semantically equal to the given attribute. The introduced conceptual modeling grammar is implemented as an artifact to store and manage attribute lattices, to graphically represent them, and integrate lattices from various heterogeneous sources. With the ever-increasing amount of unstructured data (mostly text data) from various data sources such as social media, integrating text data with other data sources has gained considerable attention. This massive amount of data, however, makes finding the data relevant to a topic of interest a new challenge. I argue that the attribute lattice provides a robust semantic foundation to address this information retrieval challenge from unstructured data sources. Hence, a topic modeling approach based on the attribute lattice is proposed for Twitter. This topic model conceptualizes topic structure of tweets related to the domain of interest and enhances information retrieval by improving the semantic interpretability of hashtags.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.004
metaresearch head score (Gemma)0.007
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: Theoretical or conceptual
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.013
Threshold uncertainty score0.027

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0040.007
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.004
Bibliometrics0.0030.003
Science and technology studies0.0020.005
Scholarly communication0.0050.009
Open science0.0030.004
Research integrity0.0020.004
Insufficient payload (model declined to judge)0.0050.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.101
GPT teacher head0.312
Teacher spread0.210 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designTheoretical or conceptual
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2021
Admission routes1
Has abstractyes

Explore more

Same venueMemorial University Research Repository (Memorial University)Same topicSemantic Web and OntologiesFrench-language works237,207