Decision tree analysis to evaluate risks associated with lameness on dairy farms with automated milking systems
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Lameness is an endemic disorder causing health problems and production losses in the dairy cow industry. The objective of this study was to identify cow and farm-level factors associated with lameness on Automatic Milking System (AMS) farms, using decision tree analysis to assign probabilities to each input. AMS farms across Canada and Michigan were evaluated to identify the most substantial farm (i.e., stall design, bedding) and cow-level (i.e., BCS, leg injuries) factors associated with prevalence of lameness. To assess lameness, videos of cows were used, and cows with a head bob or noticeable limp were categorized as lame. A decision tree classification model used 1378 data points from 39 pens across 36 farms to predict the value of the target class through “tree function” in MATLAB. The primary classifier was identified as type of stall base, dividing the data set into 3 categories: 1) rubber, sand, or geotextile mat flooring, 2) concrete base, and 3) other types of stall base. Within the first category (class membership (CM) = 976), bedding quantity was the secondary classifier, which was divided by cows standing on ≥2 cm (CM=456) or <2 cm (CM=520) of bedding. Bedding quantity was divided into the third most important classifier of BCS, and cow fit stall width. Cows with BCS of 3.25 to 4.5 (CM=307) were defined as non-lame with an estimated probability (EP) of 0.59, while cows with BCS of 2 to 2.5 (CM=213) were further split by hock lesion incidence. Cows without lesions were defined non-lame (EP=0.93) and cows with lesions were defined lame (EP=0.07). Cows that fit stall width were defined as non-lame (EP=0.66) and cows that did not fit were further divided by the width of the feed alley. Farms with ≥430 cm feed alley were defined as non-lame (EP=0.89), whereas farms with <430 cm feed alley were defined as lame (EP=0.11). Through implementing a novel multifactorial approach of data analysis, we were able to highlight the critical points that can be focused on to enhance farm-level housing and management practices or mitigate or monitor cow-level issues to reduce incidence and severity of lameness in AMS farms.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.006 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it