Determination of the Classification Success of KNN Algorithm Distance Metric Methods on Wheat Seeds Dataset
Bibliographic record
Abstract
Machine learning algorithms are widely used in product sorting processes in the food industry. The attributes of the products are used in the classification process. Attributes vary for each product. In this study, using the k nearest neighbor (KNN) algorithm, the classification of the wheat groups of Kama, Rosa and Canada was performed. The Seeds dataset provided in UCI (University of California, Irvine) machine learning open source data storage was used. There are 70 examples of each wheat class in the data set. In addition, the classification estimation success of distance metrics and the number of training data was measured. Each of the wheat samples was randomly selected and a soft X-ray technique was used to visualize the inner core structure of the wheat in the experimental environment with high quality. According to the training rates ranging from 50% to 90% of the data set, the classification success of the KNN algorithm was tested. In the KNN algorithm, the neighborhood values 1, 3 and 5 were selected to affect the classification success. The successes of the Euclidean, Chebyshev, Manhattan and Mahalanobis distance metric methods of the KNN algorithm were tested according to each k neighborhood value. According to the results obtained, with the Mahalanobis metric method, a classification success rate of 0.9924 accuracy was obtained according to the AUC (Area Under the Curve) success metric by using the neighborhood value of k = 3. In the literature, there is no study comparing the KNN algorithm, neighborhood values and distance vectors together on food data sets using varying training and test data. Therefore, it is thought that the study will make an important contribution to the literature.
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 | 0.004 | 0.012 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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.
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".