It is to be expected that any machinery weighing in the tonnes like crawlers, dozer, tractors, and cranes, is going to create lots of noise. In many instances, this need not be too much of an issue; in mining, for example, workers can be protected from the noise with PPE, while civilization is often many miles away, unaffected by noise and any subsequent disruption. In other instances, however, proximity to residences and workplaces is unavoidable, creating public disturbance. There are a number of ways to treat this noise and minimize the impact on nearby homes and business, and any potential detrimental effects on operations.
Besides the removal or replacement of the machinery (which is difficult or impossible based on the specifications of the job), implementing engineering controls at the noise source is the most effective way of reducing exposure to noise. According to the hierarchy of hazard controls, engineering controls will isolate workers and anyone else from the noise hazard. On heavy machinery weighing many tons, engine compartments emitting excessive noise can be covered with portable and lightweight Echo Barriers. With reductions of 10 to as much as 30 decibels achievable in the field, implementing Echo Barriers is one of the quickest and surest ways of mitigating heavy equipment noise.
Although not as effective as an isolation approach, installing Echo Barriers around a site perimeter or a troublesome area is useful for minimizing the noise escaping beyond the worksite. Additionally, Echo Barriers are effective at helping communicate a positive corporate message to those who see them, informing residents that an active effort is being made to mitigate noise impact.
Typical noise exposure levels vary between different construction trades. Crane and bulldozer operators are reported to be exposed more than many other tradesmen such as carpenters and heating, ventilation, and air conditioning installers, with average exposure levels measuring 93 to 105 decibels. While concern should be shown for broader environmental impact of operational noise, the workers health shouldnt be neglected either.
Echo Barrier is an innovative portable noise control system designed to mitigate noise in dynamic and sensitive work environments,reducing noise by up to 43 dB. These portable noise barriers can be used in conjunction with personal protective equipment to maximize hearing protection for both workers and the community.
At Kennametal, we are serious about mine safety. We know dust and noise are pervasive safety hazards underground and if they arent controlled, can result in worker injury, lost production, and government fines.
Kennametals ProPoint bits are slim, carbide-tipped, and designed to cut coal with better penetration and far less dust. A slimmer tool produces less dust because it cuts the coal more cleanly into larger chunks, rather than pulverizing it.
Larger coal chunks are easier to sort, process, and sell. Processing coal fines, or grit, for sale is expensive, and so is disposing of it. Because ProPoint penetrates better and reduces dust, it can reduce your processing costs, too.
Noise sources in roof drilling are centered about 4 to 8 inches below the drill steel/roof intersection and above the drill chuck. As the drill steel moves during cutting, the chuck drill steel noise moves, too. This creates an intense level of noise for the drill operator.
Underground equipment operators cannot be exposed to noise levels beyond government-defined limits during their shifts. If that happens, the operators are removed from the work area and production is stopped in that mine section. That can be costly.
The drill steel-isolator-drill steel sequence isolates the vibration and noise that travel the length of the drill steel. Testing of our bit isolator has shown a noise reduction of 3-8 db in drilling operations.
In the real world, noisy data brings tremendous challenges to data mining. Traditional classification methods are proven to be inadequate to assess the efficacy of the data mining methods while using noisy and imbalanced data. Therefore, preprocessing the imbalanced data is necessary before classification. But it's difficult to arrive at an appropriate classifier for minority class in the imbalanced data. This paper proposes the hybridization of two techniques, Noise reduction and oversampling techniques which only oversamples or strengthens the borderline minority class. The proposed technique is applied on 49 datasets at several imbalanced ratios. The Decision Tree, Gaussian Naive Bayes, Logistic Regression, Neural Network, Non-linear SVM, Random Forest, and SVM using Linear Kernel classifiers are applied for getting validation through experiments. These experimental outputs show the proposed oversampling method is superior giving accurate results in imbalanced data than the random oversampling approach.
Yin, Q.-Y., Zhang, J.-S., Zhang, C.-X., & Liu, S.-C. (2013). An empirical study on the performance of cost-sensitive boosting algorithms with different levels of class imbalance. Mathematical Problems in Engineering, vol. 2013. Article ID, 761814, 112. https://doi.org/10.1155/2013/761814
Stefanowski, J. (2016). Dealing with data difficulty factors while learning from imbalanced data. In S. Matwin & J. Mielniczuk (Eds.), Challenges in computational statistics and data mining (vol. 605, pp. 333363).
Zadrozny, B., & Elkan, C. (2001). Learning and making decisions when costs and probabilities are both unknown. Proceedings of the Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, . https://doi.org/10.1145/502512.502540
Domingos, P. (1999, August). Metacost: A general method for making classifiers cost-sensitive. InProceedings of the fifth ACM SIGKDD international conference on knowledge discovery and data mining(pp. 155164).
He, H., Bai, Y., Garcia, E. A., & Li, S. (2008, June). ADASYN: Adaptive synthetic sampling approach for imbalanced learning. In2008 IEEE international joint conference on neural networks (IEEE world congress on computational intelligence)(pp. 13221328). IEEE.
Krawczyk, B., Woniak, M., & Herrera, F. (2015). On the usefulness of one-class classifier ensembles for decomposition of multi-class problems. Pattern Recognition,48(12), 39693982. https://doi.org/10.1016/J.PATCOG.2015.06.001
Wang, S., Li, Z., Chao, W., & Cao, Q. (2012, June). Applying adaptive over-sampling technique based on data density and cost-sensitive SVM to imbalanced learning. InThe 2012 international joint conference on neural networks (IJCNN) (pp. 18). IEEE.
Fernandez, A., Garcia, S., Herrera, F., & Chawla, N. V. (2018). SMOTE for learning from imbalanced data: Progress and challenges, marking the 15-year anniversary. Journal of Artificial Intelligence Research, 61, 863905.
Rivera, W. A., Goel, A., & Kincaid, J. P. (2014, December). OUPS: a combined approach using SMOTE and Propensity Score Matching. In2014 13th international conference on machine learning and applications (pp. 424427). IEEE.
Fernndez, A., Lpez, V., Galar, M., Del Jesus, M. J., & Herrera, F. (2013). Analysing the classification of imbalanced data-sets with multiple classes: Binarization techniques and ad-hoc approaches.Knowledge-Based Systems,42, 97110.
Laurikkala, J. (2001). Improving identification of difficult small classes by balancing class distribution. InConference on artificial intelligence in medicine in Europe (pp. 6366). Berlin, Heidelberg: Springer.
Zhang, J. P., & Mani, I. (2003). KNN approach to unbalanced data distributions: A case study involving information extraction. In Proceedings of international conference on machine learning (ICML 2003), workshop on learning from imbalanced data sets, Washington DC.
Ramentol, E., Caballero, Y., Bello, R., et al. (2012). SMOTE-RSB: A hybrid preprocessing approach based on oversampling and undersampling for high imbalanced data-sets using SMOTE and rough sets theory. Knowledge and Information Systems, 33, 245265. https://doi.org/10.1007/s10115-011-0465-6
Ramentol, E., Verbiest, N., Bello, R., Caballero, Y., Cornelis, C., & Herrera, F. (2012). SMOTE-FRST: A new resampling method using fuzzy rough set theory. InUncertainty modeling in knowledge engineering and decision making (pp. 800805).
Bunkhumpornpat, C., Sinapiromsaran, K., & Lursinsap, C. (2009, April). Safe-level-smote: Safe-level-synthetic minority over-sampling technique for handling the class imbalanced problem. InPacific-Asia conference on knowledge discovery and data mining(pp. 475482). Springer.
Han, H., Wang, W. Y., & Mao, B. H. (2005, August). Borderline-SMOTE: A new over-sampling method in imbalanced data sets learning. InInternational conference on intelligent computing(pp. 878887). Springer.
Barua, S., Islam, M. M., Yao, X., & Murase, K. (2014). MWMOTE Majority weighted minority oversampling technique for imbalanced data set learning. IEEE Transactions on Knowledge and Data Engineering, 26, 405425.
RanjaniRani, R., & Ramyachitra, D. (2018). Microarray cancer gene feature selection using spider monkey optimization algorithm and cancer classification using svm. Procedia Computer Science, 143, 108116. https://doi.org/10.1016/j.procs.2018.10.358
Priya, V. S., & Ramyachitra, D. (2019). Modified genetic algorithm (MGA) based feature selection with mean weighted least squares twin support vector machine (MW-LSTSVM) approach for vegetation classification. Cluster Comput, 22, 1356913581.
William, A. R., & Xanthopoulos, P. (2016). A priori synthetic over-sampling methods for increasing classification, sensitivity in imbalanced data sets. Expert Systems with Applications, 66, 124135. https://doi.org/10.1016/j.eswa.2016.09.010
Austin, P. C. (2011). An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behavioral Research, 46(3), 399424. https://doi.org/10.1080/00273171.2011.568786
Pankajavalli, P. B., & Karthick, G. S. (2020). A unified framework for stress forecasting using machine learning algorithms. In R. Chillarige, S. Distefano & S. Rawat (Eds.), Advances in computational intelligence and informatics. ICACII 2019. Lecture Notes in Networks and Systems, 119. Singapore: Springer.
Pavithra, P., Pankajavalli, P. B., & Karthik, G. S. (2019). Iot-based non-invasive breath analysis using bagged decision tree for prediction and classification of diabetes mellitus.Journal of Advanced Research in Dynamical and Control Systems, 11(06 - Special Issue), 13771382.
Existing deep learning models applied to reaction prediction in organic chemistry can reach high levels of accuracy (>90% for natural language processing-based ones). With no chemical knowledge embedded other than the information learnt from reaction data, the quality of the datasets plays a crucial role in the performance of the prediction models. Human curation is prohibitively expensive, so unaided approaches to remove chemically incorrect entries from existing datasets are essential to improve the performance of artificial intelligence models in synthetic chemistry tasks. Here, we propose a machine learning-based, unassisted approach to remove chemically wrong entries from chemical reaction collections. We apply this method to the Pistachio collection of chemical reactions and to an open dataset, both extracted from United States Patent and Trademark Office patents. Our results show an improved prediction quality for models trained on the cleaned and balanced datasets. For retrosynthetic models, the roundtrip accuracy metric grows by 13 percentage points and the value of the cumulative JensenShannon divergence decreases by 30% compared to its original record. The coverage remains high at 97%, and the value of the class diversity is not affected by the cleaning. The proposed strategy is the first unassisted rule-free technique to address automatic noise reduction in chemical datasets.
The data that support the findings of this study are the reaction dataset Pistachio 3 (version release of 18 November 2019) from NextMove Software3. It is derived by text-mining chemical reactions in US patents. We also used two smaller open-source datasets: the dataset by Schneider et al.34, which consists of 50,000 randomly picked reactions from US patents, and the USPTO dataset by Lowe2, an open dataset with chemical reactions from US patents (1976 to September 2016). A demonstration of the code on the dataset by Schneider et al. is also available in the GitHub repository (https://github.com/rxn4chemistry/OpenNMT-py/tree/noise_reduction). Source data for the plots in the main manuscript are available at https://figshare.com/articles/journal_contribution/Source_Data/13674496.
Schwaller, P. & Laino, T. Data-Driven Learning Systems for Chemical Reaction Prediction: An Analysis of Recent Approaches. In Machine Learning in Chemistry: Data-Driven Algorithms, Learning Systems and Predictions (eds. Pyzer-Knapp, E. O. & Laino, T.) 6179 (ACS Publications, 2019).
Thakkar, A., Kogej, T., Reymond, J. L., Engkvist, O. & Esben, J. Datasets and their influence on the development of computer assisted synthesis planning tools in the pharmaceutical domain. Chem. Sci 11, 154168 (2020).
Dai, H., Li, C., Coley, C., Dai, B. & Song, L. Retrosynthesis prediction with conditional graph logic network. In Proc. Advances in Neural Information Processing Systems 32 (eds Wallach, H. et al.) 88728882 (Curran Associates, 2019).
Sacha, M., Ba, M., Byrski, P., Wodarczyk-Pruszyski, P. & Jastrzebski, S. Molecule edit graph attention network: modeling chemical reactions as sequences of graph edits. Preprint at https://arxiv.org/pdf/2006.15426.pdf (2020).
Nguyen, H. V. & Vreeken, J. Non-parametric JensenShannon divergence. In Machine Learning and Knowledge Discovery in Databases. ECML PKDD 2015. Lecture Notes in Computer Science Vol. 9285 (eds. Appice, A. et al.) 173189 (Springer, 2015).
A.T. and P.S. conceived the idea and performed the experiments. A.T. verified the statistical results of the method. A.C. carried out the chemical evaluation of the results. J.G. helped with the software implementation. T.L. supervised the project. All authors participated in discussions and contributed to the manuscript.
a, Statistical experiment: detecting whether the selection of most forgotten examples by the retro model is random. b, Overlap between the data sets removed by forward and retro models. The red line models how the overlap would be if the retro selection were entirely random. c, Percentage of one-precursor reactions in the data set cleaned by the forward forgetting and by the retro forgetting models. The former is able to identify them, while the latter is not.
Toniato, A., Schwaller, P., Cardinale, A. et al. Unassisted noise reduction of chemical reaction datasets. Nat Mach Intell 3, 485494 (2021). https://doi.org/10.1038/s42256-021-00319-wGet in Touch with Mechanic