| 1. |
- Bashir, Sarmad, et al.
(författare)
-
Requirement or Not, That is the Question : A Case from the Railway Industry
- 2023
-
Ingår i: <em>Lecture Notes in Computer Science. </em>Volume 13975. Pages 105 - 121 2023. - : Springer Science and Business Media Deutschland GmbH. - 9783031297854 ; , s. 105-121
-
Konferensbidrag (refereegranskat)abstract
- Requirements in tender documents are often mixed with other supporting information. Identifying requirements in large tender documents could aid the bidding process and help estimate the risk associated with the project. Manual identification of requirements in large documents is a resource-intensive activity that is prone to human error and limits scalability. This study compares various state-of-the-art approaches for requirements identification in an industrial context. For generalizability, we also present an evaluation on a real-world public dataset. We formulate the requirement identification problem as a binary text classification problem. Various state-of-the-art classifiers based on traditional machine learning, deep learning, and few-shot learning are evaluated for requirements identification based on accuracy, precision, recall, and F1 score. Results from the evaluation show that the transformer-based BERT classifier performs the best, with an average F1 score of 0.82 and 0.87 on industrial and public datasets, respectively. Our results also confirm that few-shot classifiers can achieve comparable results with an average F1 score of 0.76 on significantly lower samples, i.e., only 20% of the data. There is little empirical evidence on the use of large language models and few-shots classifiers for requirements identification. This paper fills this gap by presenting an industrial empirical evaluation of the state-of-the-art approaches for requirements identification in large tender documents. We also provide a running tool and a replication package for further experimentation to support future research in this area. © 2023, The Author(s)
|
|
| 2. |
|
|
| 3. |
|
|
| 4. |
- Bohlin, Pernilla, 1976, et al.
(författare)
-
Observations on persistent organic pollutants in indoor and outdoor air using passive polyurethane foam samplers
- 2008
-
Ingår i: Atmospheric Environment. - : Elsevier BV. - 1352-2310. ; 42:31, s. 7234-7241
-
Tidskriftsartikel (refereegranskat)abstract
- Air quality data of persistent organic pollutants (POPs) indoors and outdoors are sparse or lacking in several parts of the world, often hampered by the cost and inconvenience of active sampling techniques. Cheap and easy passive air sampling techniques are therefore helpful for reconnaissance surveys. As a part of the Megacity Initiative: Local and Global Research Observations (MILAGRO) project in Mexico City Metropolitan Area in 2006, a range of POPs (polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), organochlorine pesticides (OCPs), and polybrominated diphenyl ethers (PBDEs)) were analyzed in polyurethane foam (PUF) disks used as passive samplers in indoor and outdoor air. Results were compared to those from samplers deployed simultaneously in Gothenburg (Sweden) and Lancaster (United Kingdom). Using sampling rates suggested in the literature, the sums of 13 PAHs in the different sites were estimated to be 6.1–180 ng m−3, with phenanthrene as the predominant compound. Indoor PAH levels tended to be higher in Gothenburg and outdoor levels higher in Mexico City. The sum of PCBs ranged 59–2100 ng m−3, and seemed to be highest indoors in Gothenburg and Lancaster. PBDE levels (sum of seven) ranged 0.68–620 ng m−3, with the highest levels found in some indoor locations. OCPs (i.e. DDTs, HCHs, and chlordanes) were widely dispersed both outdoors and indoors at all three studied areas. In Gothenburg all POPs tended to be higher indoors than outdoors, while indoor and outdoor levels in Mexico City were similar. This could be due to the influence of indoor and outdoor sources, air exchange rates, and lifestyle factors. The study demonstrates how passive samplers can provide quick and cheap reconnaissance data simultaneously at many locations which can shed light on sources and other factors influencing POP levels in air, especially for the gaseous fractions.
|
|
| 5. |
|
|
| 6. |
- Bohlin, Pernilla, 1976
(författare)
-
Passive sampling of PAHs and some trace organic compounds in occupational and residential air: needs, evaluation and limits
- 2010
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Human exposure to elevated levels of polycyclic aromatic hydrocarbons (PAHs) and trace organic compounds (e.g. PCB, OCP, PBDE) can be related to negative health effects. This raises demands of exposure measurements to be performed. Passive air samplers (PAS) are simple and cheap sampling options that may be useful tools for exposure screening of large populations. They can also enable frequent monitoring. The aim of this thesis was to evaluate and increase the understanding of PAS methods to sample and monitor PAHs and some trace organic compounds in occupational and residential air. A polyurethane foam (PUF) based PAS was considered having the best potential and was chosen for the evaluation. Two designs of this PUF-PAS were evaluated: one for stationary sampling shaped as a disk and one for personal sampling shaped as a cylinder (“mini-PUF”). The results presented show that PUF-PAS disks and mini-PUFs provide detectable levels for most of the studied compounds under typical occupational and residential air concentrations. They also showed potential to detect spatial differences in concentrations between and within sampling sites as well as inside and outside homes. The novel use of the mini-PUF was successful both as stationary and personal sampler. Moreover, the precision of gas phase PAHs in replicates of PUF-PAS disks and mini-PUFs were comparable to precision for active samplers while particle-associated PAHs showed more variable results. Results from personally deployed mini-PUFs were significantly correlated to personal active samplers for the studied compounds and the accuracy was high for most compounds. Sampling rates (RS) for 16 individual PAHs ranged from 1 to 10 m3 day-1 (0.7-7 L min-1) in PUF-PAS disks, from 0.4 to 3.3 m3 day-1 (0.3-2.3 L min-1) in mini-PUFs deployed for two weeks as well as 8 h. No significant differences in RS were found for PAHs in the gas phase and PAHs associated to particles. The RS was higher for the mini-PUF compared to the PUF-PAS disk when correcting for their surface areas indicating a more efficient uptake in the mini-PUF design. Somewhat higher RS was also found for gas phase PAHs for the 8 h exposure compared to two weeks exposure. In conclusion, this thesis demonstrates that PUF-PAS disks can be a useful tool for screening of PAH concentrations in occupational environments. The mini-PUF has a good potential to be used as a personal sampler for PAHs in occupational environments but requires further validation.
|
|
