Abstract
Existing researcher-level metrics typically collapse a scholar's entire academic footprint into a single citation-derived number. Such indicators, while convenient, fail to capture the breadth of scholarly activity, penalize early-career researchers, and remain vulnerable to citation manipulation. The Orclever Influence Score (OIS) addresses these limitations by integrating five orthogonal dimensions of academic contribution into a single composite indicator on a 0–100 scale. Each dimension is independently normalized, weighted, and aggregated, providing a more balanced and interpretable signal of scholarly influence.
1.Introduction
The evaluation of scholarly impact is a persistent challenge in research assessment. Single-metric approaches—while widely adopted—suffer from known deficiencies: they overweight citation accumulation, favor established researchers in citation-rich disciplines, and ignore non-publication contributions such as peer review, editorial stewardship, and cross-disciplinary engagement.
OIS was developed to provide a more holistic signal. Rather than replacing existing indicators, it synthesizes multiple evidence streams into a composite score that captures not only what a researcher has produced, but how that work was verified, when it was published, how broadly the researcher contributes, and what service they render to the scholarly ecosystem.
The design philosophy follows three principles: (i) all inputs must be automatically verifiable from authoritative metadata records; (ii) the model must degrade gracefully when data is incomplete; and (iii) the scoring must remain transparent and reproducible.
2.Model Architecture
OIS is defined as a weighted linear combination of five independently normalized sub-scores, each bounded to the [0, 100] interval:
The weight allocation reflects a deliberate hierarchy: citation-based influence receives the largest share (35%), followed by verified output quality (25%), ensuring that the core academic signal—peer-recognized research—dominates the composite. Research momentum (20%) prevents lifetime-accumulated advantages from masking current inactivity. Contribution breadth (10%) and scholarly service (10%) reward dimensions of academic life that citation counts alone cannot capture.
| Code | Component | Weight |
|---|---|---|
| NI | Normalized Influence | 0.35 |
| VQ | Verified Output Quality | 0.25 |
| RM | Research Momentum | 0.20 |
| CB | Contribution Breadth | 0.10 |
| SS | Scholarly Service | 0.10 |
3.Normalized Influence (NI)
3.1 Rationale
Raw citation counts are inadequate proxies for influence: they conflate self-citations with independent recognition, scale nonlinearly across disciplines, and are disproportionately inflated by a small number of highly-cited works. NI addresses these issues through three mechanisms: self-citation exclusion, logarithmic compression, and h-index integration.
3.2 Computation
Let Cind denote the total number of independent citations (excluding author self-citations) across all of a researcher's publications. When independent citation data is unavailable, a conservative estimate is derived by applying a 15% self-citation discount to the gross citation total.
The logarithmic transformation ensures diminishing returns at high citation volumes—the difference between 50 and 500 citations is weighted more heavily than between 5,000 and 50,000—reducing the dominance of outlier publications. The citation component contributes up to 70 points.
An additional h-index bonus captures the consistency of a researcher's citation distribution rather than total volume:
This dual structure ensures that a researcher with modest total citations but a consistently cited body of work (high h-index) is not penalized relative to one with a single highly-cited paper.
4.Verified Output Quality (VQ)
4.1 Rationale
Not all publications carry equal epistemic weight. A peer-reviewed journal article with a persistent identifier represents a fundamentally different contribution than a self-declared conference abstract without verifiable metadata. VQ assigns differentiated credit based on three multiplicative factors: publication type, verification strength, and venue quality.
4.2 Computation
For each publication i, a composite quality score is computed:
Where:
- τi is the publication type weight: review articles (1.10), journal articles (1.00), books and monographs (0.90), proceedings papers (0.75), book chapters (0.70), dissertations (0.60), editorials (0.50), preprints (0.30).
- νi is the verification coefficient: DOI-linked and publisher-verified records receive full credit (1.00); OAuth-verified records (0.90); platform-verified (0.85); self-claimed entries are discounted (0.60).
- γi is the venue quality multiplier, derived from the publishing venue's two-year mean citedness—a normalized indicator of journal-level citation performance. This multiplier ranges from 1.00 (baseline) to 1.50 (high-impact venues).
- αi is the author position weight: sole authorship (1.20×), first authorship (1.10×), last/senior authorship (1.05×), and middle authorship (0.85×). Author position is determined algorithmically from the ordered contributor list in the publication metadata.
The scaling constant S = 15 is calibrated such that approximately 15 well-verified, first-authored journal articles in reputable venues yield a maximum score. This threshold aligns with the typical mid-career output of an active researcher.
5.Research Momentum (RM)
5.1 Rationale
A researcher's cumulative record may reflect past productivity without indicating current activity. RM measures the recency of scholarly engagement, ensuring that active researchers are appropriately recognized and that legacy metrics do not indefinitely inflate scores.
5.2 Computation
RM consists of two sub-components measured over a rolling three-year window:
Publication momentum (up to 60 points) measures the proportion of a researcher's output published in the last three years relative to their expected activity rate. A researcher who has published 5 papers in the last three years out of a career total of 10 receives a higher momentum score than one with 5 recent papers out of 100.
Citation momentum (up to 40 points) captures whether a researcher's work is gaining or losing citation traction. It compares recent-year citations to the researcher's own career average, rewarding those whose work is currently resonating with the community.
6.Contribution Breadth (CB)
6.1 Rationale
Scholarly influence extends beyond any single mode of contribution. A researcher who publishes articles, reviews manuscripts for journals, serves on editorial boards, and engages across multiple disciplines demonstrates a form of academic leadership that mono-dimensional metrics cannot detect. CB quantifies this diversity.
6.2 Computation
CB is composed of four equally-weighted diversity indicators, each contributing up to 25 points:
- Output type diversity: the number of distinct publication formats (journal articles, book chapters, proceedings, reviews, etc.) across the researcher's portfolio.
- Research field diversity: the number of declared research specializations, reflecting interdisciplinary engagement.
- Collaboration breadth: the number of unique co-authors, indicating network reach and collaborative capacity.
- Role diversity: whether the researcher has contributed as an author, a peer reviewer, and an editor—the three pillars of the scholarly publishing ecosystem.
7.Scholarly Service (SS)
7.1 Rationale
The infrastructure of peer review depends on researchers who dedicate time to evaluating manuscripts, curating journal content, and maintaining editorial standards. These contributions, though essential to scientific integrity, receive no recognition in citation-based metrics. SS corrects this asymmetry.
7.2 Computation
- Peer review activity (up to 40 points): each completed review contributes 8 points, reflecting the substantial time and expertise required for rigorous evaluation.
- Editorial roles (up to 35 points): serving as an editor or section editor for a journal contributes 15 points per role, acknowledging the sustained commitment to field stewardship.
- Editor-in-Chief service (10-point bonus): recognizes the highest level of editorial responsibility, including strategic direction and quality assurance.
- Identity verification (up to 15 points): verified ORCID (5), verified institutional affiliation (5), and verified email (5) collectively signal a researcher's commitment to transparency and accountability in the scholarly record.
8.Score Interpretation
OIS maps the composite score to six interpretive levels, providing an intuitive summary of a researcher's current academic standing:
| Score Range | Interpretation |
|---|---|
| 85 – 100 | Exceptional Influence |
| 70 – 84 | Strong Influence |
| 55 – 69 | Established Influence |
| 40 – 54 | Emerging Influence |
| 20 – 39 | Growing Influence |
| 0 – 19 | Early-Stage |
These labels are descriptive, not normative. An “Early-Stage” designation does not imply inferior quality—it indicates that the researcher's publicly verifiable evidence base is still developing. As publications accumulate, citations grow, and service contributions are recorded, the score will naturally evolve.
9.Design Principles & Limitations
9.1 Transparency
Every component of OIS is deterministic and reproducible. Given the same input data, the same score will always be produced. There are no opaque machine-learning models, no proprietary weighting adjustments, and no editorial overrides. The complete scoring algorithm is published in this document.
9.2 Graceful Degradation
OIS is engineered to produce meaningful scores even when external metadata is incomplete. If a researcher has no citation data, the citation component contributes zero—but the output quality, momentum, breadth, and service dimensions remain fully functional. This ensures that early-career researchers or those in emerging fields are not rendered invisible.
9.3 Anti-Manipulation Safeguards
Self-citations are explicitly separated from independent citations. The logarithmic compression of citation counts reduces the impact of citation inflation. Verification-weighted output quality discounts unverified claims. Author position weighting prevents honorary authorship from inflating scores.
9.4 Limitations
OIS does not claim to measure research quality in an absolute sense. It does not normalize across disciplines—a computer scientist and a humanities scholar with identical OIS scores may have fundamentally different publication and citation profiles. The score is best interpreted as a relative signal within a researcher's own trajectory, not as a cross-disciplinary ranking tool.
10.Conclusion
The Orclever Influence Score represents a deliberate departure from single-metric reductionism in research assessment. By integrating citation impact, output verification, temporal dynamics, contribution diversity, and scholarly service into a unified framework, OIS provides a more complete—though necessarily imperfect—portrait of a researcher's academic influence.
The score is intended as an interpretive tool, not a definitive judgment. It surfaces patterns that individual metrics cannot detect: the early-career researcher building momentum, the senior scholar whose editorial service sustains an entire field, the interdisciplinary thinker whose breadth defies narrow categorization.
OIS will continue to evolve. Future iterations may incorporate field-normalized citation benchmarks, institutional prestige adjustments, and additional contribution types as the scholarly communication landscape develops.