The majority of the machine-learning students believed their presentation skills to be average (41%), good (25%) or even excellent (18%). Only a small proportion rated their skills as poor (9%) or very low (7%).
We published a proposal on ResearchGate (DOI: 10.13140/RG.2.2.18759.39842) for a novel approach to the calculate relatedness of research articles and other document types. The relatedness of research articles, patents, legal documents, web pages, and other documents is often calculated with citation or hyperlink based approaches such as citation proximity analysis (CPA). Read more…
According to the students´ self-assessment, they had – on average – average knowledge of JAVA (3.2 out of 5 points). However, there was quite a bit of variance. Many students had very little knowledge of JAVA, while others had excellent knowledge. The knowledge of C/C++/C# was similar to JAVA (3.1 on Read more…
The top careers for the machine-learning students were machine-learning engineer (16%), software engineer (16%), business start-up founder (14%), and data scientist (13%). However, also careers in project management (10%), research (9%) and consultancy (8%) were interesting to some students. “Business manager” was the least attractive career to the machine-learning students. Read more…
On average, the machine-learning students had average knowledge in databases and general writing (around 3 out of 5 points), below-average knowledge in writing research papers (1.9 out of 5 points), and little knowledge in machine learning (1.5 out of 5). The little knowledge about machine learning is not surprising though, Read more…
10% of the students had very low motivation to participate in the module, while the majority of the students had high (28%) or even very high (48%) motivation to learn about machine learning.
We published a preprint on arXiv, in which we question the current practice of calculating evaluation metrics for recommender systems as single numbers (e.g. precision p=.28 or mean absolute error MAE = 1.21). We argue that single numbers express only average effectiveness over usually a rather long period (e.g. a Read more…
In the coming term, Prof. Douglas Leith and I will be teaching an introduction to machine learning here at Trinity College Dublin, Ireland. We expect around 200 students, primarily from the new Master of Science in Intelligent System and Data Science. Many students will already have a background in machine Read more…
On 28th February, we released version 1.1.1 of Mr. DLib’s recommender system with some minor improvements and bug fixes: Improved 404 error handling for unknown document IDs Fix: The order of authors in the XML was not sorted properly Several internal changes (adjusted logging table; click time is not updated any Read more…
Six months ago, we launched Mr. DLib’s recommendations-as-a-service for Academia. Time, to look back and provide some numbers: Since September 2016, Mr. DLib´s recommender system has delivered 60,836,800 recommendations to our partner Sowiport, and Sowiport’s users have clicked 91,545 of the recommendations. This equals on overall click-through rate (CTR) of Read more…
This semester my colleague and I are teaching the Machine-Learning module CS7CS4 at Trinity College Dublin. We decided to publish the lecture slides on Slideshare. The slides contain 11 lectures about machine learning, covering aspects such as an introduction to machine learning, standard machine learning algorithms (regression, support vector machines, k-means, Read more…
Two of our papers about weighting citations and terms in the context of user modeling and recommender systems got accepted at the iConference 2017. Here are the abstracts, and links to the pre-print versions: Evaluating the CC-IDF citation-weighting scheme: How effectively can ‘Inverse Document Frequency’ (IDF) be applied to references? In Read more…
