Detecting and Imputing Hidden Missing Values in Time Series Data : Case study: Alfa Laval
Bibliographic record
Abstract
Although identifying missing values in regular time series is trivial,detecting them becomes a challenge with irregular timestamps. Toreduce the storage, our partner, Alfa Laval, uses an engineering trickto store measurements in time series databases only when their valuechanges. This solution, despite solving storage problems, can createproblems in data analysis. It also complicates the identification ofmissing values. We address two problems: identifying hidden missing values fromirregular time series and developing effective imputation techniquesfor them. We use a rule-based approach to locate hidden missing val-ues tailored to the Alfa Laval dataset. Once we have identified the po-sition of hidden missing values, imputing them becomes the greaterchallenge, particularly when missing gaps are long. Our experimentsshow that while Linear Interpolation often outperforms LSTM andARIMA, it only creates a straight line between two points, failing tocapture the shape of the missing data. Consequently, in long-termgaps, we miss lots of informative fluctuations. To address these limitations, we employ a pattern-based similar-ity search method, which effectively captures the value and shape oftime series data for more accurate imputation. This thesis presentsour novel approach, which we validate on a subset of Alfa Laval’ssensor data and three additional external datasets, demonstrating itsgeneralizability and effectiveness. While the rule-based identificationtechnique is particularly relevant to Alfa Laval’s data, our imputationtechnique serves as a general solution for time series imputation
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 | 0.002 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".