A graph digital signal processing method for semantic analysis
Bibliographic record
Abstract
This paper focuses on the problem of devising a computationally tractable procedure for representing the natural language understanding (NLU). It approaches this goal, by using distributional models of meaning through a method from graph-based digital signal processing (DSP) which only recently grabbed the attention of researchers from the field of natural language processing (NLP) related to big data analysis. The novelty of our approach lies in the combination of three domains: advances in deep learning algorithms for word representation, dependency parsing for modeling inter-word relations and convolution using orthogonal Hadamard codes for composing the two previous areas, generating a unique representation for the sentence. Two types of problems are resolved in a new unified way: sentence similarity given by the cos function of the corresponding vectors and question-answering where the query is matched to possible answers. This technique resembles the spread spectrum methods from telecommunication theory where multiple users share a common channel, and are able to communicate without interference. In the content of this paper the case of individual words play the role of users sharing the same sentence. Examples of the method application to a standard set of sentences, used for benchmarking the accuracy and the execution time is also given.
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 imitationNot 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.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.014 | 0.008 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".