ZoekresultatenZoekresultaten
Volg ons op
Volg ons op
Types
0Instelling
26Bestandstype
9Taal
5Publicatiejaar
12Thema's
14Producttype
14Publicaties met bestand / URL
2Projectstatus
3Uit de aankondiging: "Steeds meer systemen loggen gegevens over hoe het bedrijfsproces verloopt, maar loopt het proces wel zoals het bedoeld was? Wat zijn de knelpunten? Text mining is vaak lastig doordat er tijdstippen ontbreken, process mining kan niet werken zonder tijdstippen, de combinatie van die twee technieken kan elkaar versterken. Bij sentiment mining weet je wel wat iemands zijn gevoelens zijn, maar niet zijn drijfveren, terwijl drijfveren juist een betere verklaring voor iemands gedrag vormen. De combinatie van deze technieken biedt mogelijkheden om nieuwe inzichten te verwerven rond customer journeys, zodat de klant uiteindelijk beter geholpen wordt." http://www.naf.nl/events/proces-text-mining/
DOCUMENT
This method paper presents a template solution for text mining of scientific literature using the R tm package. Literature to be analyzed can be collected manually or automatically using the code provided with this paper. Once the literature is collected, the three steps for conducting text mining can be performed as outlined below:
• loading and cleaning of text from articles,
• processing, statistical analysis, and clustering, and
• presentation of results using generalized and tailor-made visualizations.
The text mining steps can be applied to a single, multiple, or time series groups of documents.
References are provided to three published peer reviewed articles that use the presented text mining methodology. The main advantages of our method are: (1) Its suitability for both research and educational purposes, (2) Compliance with the Findable Accessible Interoperable and Reproducible (FAIR) principles, and (3) code and example data are made available on GitHub under the open-source Apache V2 license.
DOCUMENT
Process Mining can roughly be defined as a data-driven approach to process management. The basic idea of process mining is to automatically distill and to visualize business processes using event logs from company IT-systems (e.g. ERP, WMS, CRM etc.) to identify specific areas for improvement at an operational level. An event log can be described as a database entry that signifies a specific action in a software application at a specific time. Simple examples of these actions are customer order entries, scanning an item in a warehouse, and registration of a patient for a hospital check-up.
Process mining has gained popularity in the logistics domain in recent years because of three main reasons. Firstly, the logistics IT-systems' large and exponentially growing amounts of event data are being stored and provide detailed information on the history of logistics processes. Secondly, to outperform competitors, most organizations are searching for (new) ways to improve their logistics processes such as reducing costs and lead time. Thirdly, since the 1970s, the power of computers has grown at an astonishing rate. As such, the use of advance algorithms for business purposes, which requires a certain amount of computational power, have become more accessible.
Before diving into Process Mining, this course will first discuss some basic concepts, theories, and methods regarding the visualization and improvement of business processes.
MULTIFILE
In this short paper, we address the interpretability of hidden layer representations in deep text mining: deep neural networks applied to text mining tasks. Following earlier work predating deep learning methods, we exploit the internal neural network activation (latent) space as a source for performing k-nearest neighbor search, looking for representative, explanatory training data examples with similar neural layer activations as test inputs. We deploy an additional semantic document similarity metric for establishing document similarity between the textual representations of these nearest neighbors and the test inputs. We argue that the statistical analysis of the output of this measure provides insight to engineers training the networks, and that nearest neighbor search in latent space combined with semantic document similarity measures offers a mechanism for presenting explanatory, intelligible examples to users.
DOCUMENT
In this article, we present CoPub 5.0, a publicly available text mining system, which uses Medline abstracts to calculate robust statistics for keyword co-occurrences. CoPub was initially developed for the analysis of microarray data, but we broadened the scope by implementing new technology and new thesauri. In CoPub 5.0, we integrated existing CoPub technology with new features, and provided a new advanced interface, which can be used to answer a variety of biological questions. CoPub 5.0 allows searching for keywords of interest and its relations to curated thesauri and provides highlighting and sorting mechanisms, using its statistics, to retrieve the most important abstracts in which the terms co-occur. It also provides a way to search for indirect relations between genes, drugs, pathways and diseases, following an ABC principle, in which A and C have no direct connection but are connected via shared B intermediates. With CoPub 5.0, it is possible to create, annotate and analyze networks using the layout and highlight options of Cytoscape web, allowing for literature based systems biology. Finally, operations of the CoPub 5.0 Web service enable to implement the CoPub technology in bioinformatics workflows. CoPub 5.0 can be accessed through the CoPub portal http://www.copub.org. © 2011 The Author(s).
DOCUMENT
handout van een labtalk waarin de onderzoeker enkele methoden beschrijft rond text mining, story mining: het herkennen van patronen in communicatie met klanten.
DOCUMENT
Data mining seems to be a promising way to tackle the problem of unpredictability in MRO organizations. The Amsterdam University of Applied Sciences therefore cooperated with the aviation industry for a two-year applied research project exploring the possibilities of data mining in this area. Researchers studied more than 25 cases at eight different MRO enterprises, applying a CRISP-DM methodology as a structural guideline throughout the project. They explored, prepared and combined MRO data, flight data and external data, and used statistical and machine learning methods to visualize, analyse and predict maintenance. They also used the individual case studies to make predictions about the duration and costs of planned maintenance tasks, turnaround time and useful life of parts. Challenges presented by the case studies included time-consuming data preparation, access restrictions to external data-sources and the still-limited data science skills in companies. Recommendations were made in terms of ways to implement data mining – and ways to overcome the related challenges – in MRO. Overall, the research project has delivered promising proofs of concept and pilot implementations
MULTIFILE
Objective: To annotate a corpus of randomized controlled trial (RCT) publications with the checklist items of CONSORT reporting guidelines and using the corpus to develop text mining methods for RCT appraisal. Methods: We annotated a corpus of 50 RCT articles at the sentence level using 37 fine-grained CONSORT checklist items. A subset (31 articles) was double-annotated and adjudicated, while 19 were annotated by a single annotator and reconciled by another. We calculated inter-annotator agreement at the article and section level using MASI (Measuring Agreement on Set-Valued Items) and at the CONSORT item level using Krippendorff's α. We experimented with two rule-based methods (phrase-based and section header-based) and two supervised learning approaches (support vector machine and BioBERT-based neural network classifiers), for recognizing 17 methodology-related items in the RCT Methods sections. Results: We created CONSORT-TM consisting of 10,709 sentences, 4,845 (45%) of which were annotated with 5,246 labels. A median of 28 CONSORT items (out of possible 37) were annotated per article. Agreement was moderate at the article and section levels (average MASI: 0.60 and 0.64, respectively). Agreement varied considerably among individual checklist items (Krippendorff's α= 0.06–0.96). The model based on BioBERT performed best overall for recognizing methodology-related items (micro-precision: 0.82, micro-recall: 0.63, micro-F1: 0.71). Combining models using majority vote and label aggregation further improved precision and recall, respectively. Conclusion: Our annotated corpus, CONSORT-TM, contains more fine-grained information than earlier RCT corpora. Low frequency of some CONSORT items made it difficult to train effective text mining models to recognize them. For the items commonly reported, CONSORT-TM can serve as a testbed for text mining methods that assess RCT transparency, rigor, and reliability, and support methods for peer review and authoring assistance. Minor modifications to the annotation scheme and a larger corpus could facilitate improved text mining models. CONSORT-TM is publicly available at https://github.com/kilicogluh/CONSORT-TM.
DOCUMENT
E-discovery projects typically start with an assessment of the collected electronic data in order to estimate the risk to prosecute or defend a legal case. This is not a review task but is appropriately called early case assessment, which is better known as exploratory search in the information retrieval community. This paper first describes text mining methodologies that can be used for enhancing exploratory search. Based on these ideas we present a semantic search dashboard that includes entities that are relevant to investigators such as who knew who, what, where and when. We describe how this dashboard can be powered by results from our ongoing research in the “Semantic Search for E-Discovery” project on topic detection and clustering, semantic enrichment of user profiles, email recipient recommendation, expert finding and identity extraction from digital forensic evidence.
MULTIFILE
We present a multilingual, open source system for cancer forum thread analysis, equipped with a biomedical entity tagger and a module for textual summarization. This system allows users to investi- gate textual co-occurrences of biomedical entities in forum posts, and to browse through summaries of long discussions. It is applied to a number of online cancer patient fora, including a gastro-intestinal cancer forum and a breast cancer forum. We propose that the system can serve as an extra source of information for medical hypothesis formulation, and as a facility for boosting patient empowerment.
DOCUMENT
