Deploying automated ticket router across the enterprise
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
Abstract With the recent advances in machine learning, the use of natural language processing (NLP) technology to support various business processes has been increasing. This paper discusses the use of NLP to route more than one million live client tickets annually to the appropriate service personnel in 67 support missions across IBM. Each mission supports a product family with multiple support teams, each requiring different skills for the engineers. We discuss three important aspects of such a large‐scale deployment: (i) The use of a centralized team with a common machine learning infrastructure and practices to support the entire enterprise. (ii) The processes and quality of such a deployment from the perspective of one support mission, namely, IBM's z/OS family. (iii) Careful monitoring of the deployed models to detect drifts in the routing behavior. Despite vast differences in the technical contents of the support missions, it is possible to define common processes and metrics across the enterprise, without requiring a dedicated machine learning team for each mission. In addition, we provide examples of the business policies and metrics from the perspective of the z/OS mission to demonstrate the utility of the approach and the outcome.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.005 |
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