Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal Journal of Technology Informatics and Engineering International Journal of Graphic Design
Sisi Meng
Accounting, University of Rochester, NY, USA
Arts Computer Science & IT

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
LLM-Inspired Ontology-Based Semantic Enrichment for FinTech M&A Intelligence in SEC Structured Disclosures Guanzheng Zhao; Dingyuan Zhang; Sisi Meng
Journal of Technology Informatics and Engineering Vol. 5 No. 2 (2026): AUGUST | JTIE : Journal of Technology Informatics and Engineering
Publisher : University of Science and Computer Technology

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51903/jtie.v5i2.566

Abstract

This study develops an ontology-based event-intelligence framework for FinTech merger-and-acquisition evidence in U.S. Securities and Exchange Commission (SEC) structured disclosures. Five quarterly SEC Financial Statement Data Sets from 2025Q1 through 2026Q1 contain 32,254 filings, 7,335 registrants, and 18,312,494 numerical XBRL facts; validation adds a 1,800-filing full-text Form 8-K sample and 192 FDIC events. A deterministic ontology maps XBRL tag names, labels, and documentation to acquisition, disposition, valuation, integration, risk, and payment concepts. The method is therefore LLM-inspired semantic enrichment rather than direct LLM extraction. Logistic regression, decision tree, and random forest classifiers are evaluated in three expanding forward-quarter tests with training-only preprocessing and threshold selection. A proximal protocol retains semantically close predictors, whereas a strict protocol excludes label-generating variables and deterministic descendants. Across the rolling tests, proximal logistic-regression M&A detection attains mean F1 = 0.981941, ROC-AUC = 0.999517, and average precision = 0.998315; strict performance falls to F1 = 0.735640, ROC-AUC = 0.930870, and average precision = 0.824107. FinTech M&A F1 declines from 0.921198 to 0.449503. Strict random forests yield F1 = 0.798129 for integration risk and 0.753987 for valuation signals. In independent full text, strict main-text-plus-exhibit F1 is 0.297482; among 24 automatically linked FDIC events in rolling test quarters, 9 are detected. Near-perfect scores therefore describe ontology reconstruction, not transaction-level accuracy.
From 8-K to Deal Cards: A Source-Grounded Visual Triage Framework for FinTech M&A Filings Sisi Meng; Dingyuan Zhang; Eric Li
International Journal of Graphic Design Vol. 4 No. 1 (2026): April | IJGD: International Journal of Graphic Design
Publisher : University of Science and Computer Technology

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51903/ijgd.v4i1.4001

Abstract

Form 8-K is a timely public channel through which U.S. registrants disclose material events, including acquisition agreements, completed transactions, shareholder communications, and related exhibits. The density of this filing stream creates a visual communication problem: analysts need to identify source-linked deal cues without losing the legal and temporal provenance of each record. This study presents a source-grounded deal-card workflow for FinTech M&A filing triage. The analysis used the official SEC quarterly master indexes and included 16,439 Form 8-K and 8-K/A filings in Q4 2025, together with 276,261 contextual records in the complete Q4 index. Symmetric companion-form windows were protected against quarter-boundary truncation with the last 14 days of Q3 2025 and the first 14 days of Q1 2026. A seven-day window identified 222 rule-defined FinTech M&A targets across 94 company names, or 1.35% of the Q4 corpus. In five-fold cross-validation grouped by CIK, text-only logistic triage reached F1 = 0.224 and M&A-context logistic triage reached F1 = 0.322. Integrated logistic and linear SVM configurations reproduced the rule-defined target exactly (F1 = 1.000). Because the target was constructed from the same FinTech and M&A-neighborhood cues available to those configurations, this exact agreement is interpreted as rule-consistency validation rather than predictive superiority. A structural interface audit found that the deal-card schema exposed 10 documented fields, nine visual anchors, and six configured decision cues. These author-defined attributes describe interface structure; they do not measure comprehension, trust, or review speed. The contribution is a method for organizing SEC filing evidence for subsequent visual review.
Co-Authors Dingyuan Zhang Eric Li Guanzheng Zhao