Machine Learning Algorithms in Heavy Process Manufacturing

In a global economy, manufacturers mainly compete with cost efficiency of production, as the price of raw materials are similar worldwide. Heavy industry has two big issues to deal with. On the one hand there is lots of data which needs to be analyzed in an effective manner, and on the other hand making big improvements via investments in cooperate structure or new machinery is neither economically nor physically viable. Machine learning offers a promising way for manufacturers to address both these problems as they are in an excellent position to employ learning techniques with their massive resource of historical production data. However, choosing modelling a strategy in this setting is far from trivial and this is the objective of this article. The article investigates characteristics of the most popular classifiers used in industry today. Support Vector Machines, Multilayer Perceptron, Decision Trees, Random Forests, and the meta-algorithms Bagging and Boosting are mainly investigated in this work. Lessons from real-world implementations of these learners are also provided together with future directions when different learners are expected to perform well. The importance of feature selection and relevant selection methods in an industrial setting are further investigated. Performance metrics have also been discussed for the sake of completion.

Knowledge translation in Uganda : a qualitative study of Ugandan midwives' and managers' perceived relevance of the sub-elements of the context cornerstone in the PARIHS framework

BACKGROUND: A large proportion of the annual 3.3 million neonatal deaths could be averted if there was a high uptake of basic evidence-based practices. In order to overcome this 'know-do' gap, there is an urgent need for in-depth understanding of knowledge translation (KT). A major factor to consider in the successful translation of knowledge into practice is the influence of organizational context. A theoretical framework highlighting this process is Promoting Action on Research Implementation in Health Services (PARIHS). However, research linked to this framework has almost exclusively been conducted in high-income countries. Therefore, the objective of this study was to examine the perceived relevance of the subelements of the organizational context cornerstone of the PARIHS framework, and also whether other factors in the organizational context were perceived to influence KT in a specific low-income setting. METHODS: This qualitative study was conducted in a district of Uganda, where focus group discussions and semi-structured interviews were conducted with midwives (n = 18) and managers (n = 5) within the catchment area of the general hospital. The interview guide was developed based on the context sub-elements in the PARIHS framework (receptive context, culture, leadership, and evaluation). Interviews were transcribed verbatim, followed by directed content analysis of the data. RESULTS: The sub-elements of organizational context in the PARIHS framework--i.e., receptive context, culture, leadership, and evaluation--also appear to be relevant in a low-income setting like Uganda, but there are additional factors to consider. Access to resources, commitment and informal payment, and community involvement were all perceived to play important roles for successful KT. CONCLUSIONS: In further development of the context assessment tool, assessing factors for successful implementation of evidence in low-income settings--resources, community involvement, and commitment and informal payment--should be considered for inclusion. For low-income settings, resources are of significant importance, and might be considered as a separate subelement of the PARIHS framework as a whole.