Sample of Academic Literature relating to data-driven applications in the logistics industry.
In Abdimomunov et al. we compare the challenges and opportunities of data-driven applications in logistics industry to the perspectives of logistics managers in the field. This dataset includes the inidividual academic sources as .txt files that werre used to generate the 1000 most common noun phrases in that sample of literature. The noun phrases were labeled as either 'opportunities' and 'challenges' and grouped into Themes that made up the Theoretical Framework. This Theoreitcal Framework was then compared to interviews with managers of logistics operations to systematically compare the most prevalent data topics for the logistics sector in both the academic and practical perspective.
Steps to reproduce
The list of academic sources was obtained from ScienceDirect.com using the following query: [("big data" OR "data driven") AND (logistics OR "supply chain")]. The sample of literature was filtered to include only Open Access sources as the full-text of the literature was required. It was further filtered by publication year: only sources from 2010 and later were included to ensure relevancy. The query was excuted on 2020-04-28. The individual files were than combined into a single object, which was then filtered by words/parts of speech, and parsed for noun phrases using the pattern package in Python. A frequency distribution of the noun phrases resulted in the 1000 most common noun phrases, which were then manually labelled and grouped into 'themes' - overarching topics that could be described as 'oppportunities' or 'challenges'. The text processing script, written in Python, can be found at the following repository: (https://github.com/daniyar-abdimomunov/nlp-tagging-academic-sources).