Search

Recommendation algorithms perform differently if the users, recommendation contexts, applications, and user interfaces vary even slightly. It is similarly observed in other fields, such as combinatorial problem solving, that algorithms perform differently for each instance presented. In those fields, meta-learning is successfully used to predict an optimal algorithm for each instance, to improve overall system performance. Per-instance algorithm selection has thus far been unsuccessful for recommender systems. In this paper we propose a per-instance meta-learner that clusters data instances and predicts the best algorithm for unseen instances according to cluster membership. We test our approach using 10 collaborative- and 4 content-based filtering algorithms, for varying clustering parameters, and find a significant improvement over the best performing base algorithm at alpha=0.053 (MAE: 0.7107 vs LightGBM 0.7214; t-test). We also explore the performances of our base algorithms on a ratings dataset and empirically show that the error of a perfect algorithm selector monotonically decreases for larger pools of algorithm. To the best of our knowledge, this is the first effective meta-learning technique for per-instance algorithm selection in recommender systems.

Agent-based modeling (ABM) is a powerful paradigm to gain insight into social phenomena. One area that ABM has rarely been applied is coalition formation. Traditionally, coalition formation is modeled using cooperative game theory. In this paper, a heuristic algorithm is developed that can be embedded into an ABM to allow the agents to find coalition. The resultant coalition structures are comparable to those found by cooperative game theory solution approaches, specifically, the core. A heuristic approach is required due to the computational complexity of finding a cooperative game theory solution which limits its application to about only a score of agents. The ABM paradigm provides a platform in which simple rules and interactions between agents can produce a macro-level effect without the large computational requirements. As such, it can be an effective means for approximating cooperative game solutions for large numbers of agents. Our heuristic algorithm combines agent-based modeling and cooperative game theory to help find agent partitions that are members of a games' core solution. The accuracy of our heuristic algorithm can be determined by comparing its outcomes to the actual core solutions. This comparison achieved by developing an experiment that uses a specific example of a cooperative game called the glove game. The glove game is a type of exchange economy game. Finding the traditional cooperative game theory solutions is computationally intensive for large numbers of players because each possible partition must be compared to each possible coalition to determine the core set; hence our experiment only considers games of up to nine players. The results indicate that our heuristic approach achieves a core solution over 90% of the time for the games considered in our experiment., Comment: 19 pages

Data analytics allows an analyst to gain insight into underlying populations through the use of various computational approaches, including Monte Carlo methods. This paper discusses an approach to apply Monte Carlo methods to hedonic games. Hedonic games have gain popularity over the last two decades leading to several research articles that are concerned with the necessary, sufficient, or both conditions of the existence of a core partition. Researchers have used analytical methods for this work. We propose that using a numerical approach will give insights that might not be available through current analytical methods. In this paper, we describe an approach to representing hedonic games, with strict preferences, in a matrix form that can easily be generated; that is, a hedonic game with randomly generated preferences for each player. Using this generative approach, we were able to create and solve, i.e., find any core partitions, of millions of hedonic games. Our Monte Carlo experiment generated games with up to thirteen players. The results discuss the distribution form of the core size of the games of a given number of players. We also discuss computational considerations. Our numerical study of hedonic games gives insight into the underlying properties of hedonic games., Comment: 44 pages

Automated per-instance algorithm selection often outperforms single learners. Key to algorithm selection via meta-learning is often the (meta) features, which sometimes though do not provide enough information to train a meta-learner effectively. We propose a Siamese Neural Network architecture for automated algorithm selection that focuses more on 'alike performing' instances than meta-features. Our work includes a novel performance metric and method for selecting training samples. We introduce further the concept of 'Algorithm Performance Personas' that describe instances for which the single algorithms perform alike. The concept of 'alike performing algorithms' as ground truth for selecting training samples is novel and provides a huge potential as we believe. In this proposal, we outline our ideas in detail and provide the first evidence that our proposed metric is better suitable for training sample selection that standard performance metrics such as absolute errors., Comment: 7th Workshop on Automated Machine Learning (AutoML 2020)

The effectiveness of recommender system algorithms varies in different real-world scenarios. It is difficult to choose a best algorithm for a scenario due to the quantity of algorithms available, and because of their varying performances. Furthermore, it is not possible to choose one single algorithm that will work optimally for all recommendation requests. We apply meta-learning to this problem of algorithm selection for scholarly article recommendation. We train a random forest, gradient boosting machine, and generalized linear model, to predict a best-algorithm from a pool of content similarity-based algorithms. We evaluate our approach on an offline dataset for scholarly article recommendation and attempt to predict the best algorithm per-instance. The best meta-learning model achieved an average increase in F1 of 88% when compared to the average F1 of all base-algorithms (F1; 0.0708 vs 0.0376) and was significantly able to correctly select each base-algorithm (Paired t-test; p < 0.1). The meta-learner had a 3% higher F1 when compared to the single-best base-algorithm (F1; 0.0739 vs 0.0717). We further perform an online evaluation of our approach, conducting an A/B test through our recommender-as-a-service platform Mr. DLib. We deliver 148K recommendations to users between January and March 2019. User engagement was significantly increased for recommendations generated using our meta-learning approach when compared to a random selection of algorithm (Click-through rate (CTR); 0.51% vs. 0.44%, Chi-Squared test; p < 0.1), however our approach did not produce a higher CTR than the best algorithm alone (CTR; MoreLikeThis (Title): 0.58%).

Information about the contributions of individual authors to scientific publications is important for assessing authors' achievements. Some biomedical publications have a short section that describes authors' roles and contributions. It is usually written in natural language and hence author contributions cannot be trivially extracted in a machine-readable format. In this paper, we present 1) A statistical analysis of roles in author contributions sections, and 2) Na\"iveRole, a novel approach to extract structured authors' roles from author contribution sections. For the first part, we used co-clustering techniques, as well as Open Information Extraction, to semi-automatically discover the popular roles within a corpus of 2,000 contributions sections from PubMed Central. The discovered roles were used to automatically build a training set for Na\"iveRole, our role extractor approach, based on Na\"ive Bayes. Na\"iveRole extracts roles with a micro-averaged precision of 0.68, recall of 0.48 and F1 of 0.57. It is, to the best of our knowledge, the first attempt to automatically extract author roles from research papers. This paper is an extended version of a previous poster published at JCDL 2018., Comment: arXiv admin note: substantial text overlap with arXiv:1802.01174

Many recommendation algorithms are available to digital library recommender system operators. The effectiveness of algorithms is largely unreported by way of online evaluation. We compare a standard term-based recommendation approach to two promising approaches for related-article recommendation in digital libraries: document embeddings, and keyphrases. We evaluate the consistency of their performance across multiple scenarios. Through our recommender-as-a-service Mr. DLib, we delivered 33.5M recommendations to users of Sowiport and Jabref over the course of 19 months, from March 2017 to October 2018. The effectiveness of the algorithms differs significantly between Sowiport and Jabref (Wilcoxon rank-sum test; p < 0.05). There is a ~400% difference in effectiveness between the best and worst algorithm in both scenarios separately. The best performing algorithm in Sowiport (terms) is the worst performing in Jabref. The best performing algorithm in Jabref (keyphrases) is 70% worse in Sowiport, than Sowiport`s best algorithm (click-through rate; 0.1% terms, 0.03% keyphrases).

Recommender systems in academia are not widely available. This may be in part due to the difficulty and cost of developing and maintaining recommender systems. Many operators of academic products such as digital libraries and reference managers avoid this effort, although a recommender system could provide significant benefits to their users. In this paper, we introduce Mr. DLib's "Recommendations as-a-Service" (RaaS) API that allows operators of academic products to easily integrate a scientific recommender system into their products. Mr. DLib generates recommendations for research articles but in the future, recommendations may include call for papers, grants, etc. Operators of academic products can request recommendations from Mr. DLib and display these recommendations to their users. Mr. DLib can be integrated in just a few hours or days; creating an equivalent recommender system from scratch would require several months for an academic operator. Mr. DLib has been used by GESIS Sowiport and by the reference manager JabRef. Mr. DLib is open source and its goal is to facilitate the application of, and research on, scientific recommender systems. In this paper, we present the motivation for Mr. DLib, the architecture and details about the effectiveness. Mr. DLib has delivered 94m recommendations over a span of two years with an average click-through rate of 0.12%.

We introduce the first 'living lab' for scholarly recommender systems. This lab allows recommender-system researchers to conduct online evaluations of their novel algorithms for scholarly recommendations, i.e., recommendations for research papers, citations, conferences, research grants, etc. Recommendations are delivered through the living lab's API to platforms such as reference management software and digital libraries. The living lab is built on top of the recommender-system as-a-service Mr. DLib. Current partners are the reference management software JabRef and the CORE research team. We present the architecture of Mr. DLib's living lab as well as usage statistics on the first sixteen months of operating it. During this time, 1,826,643 recommendations were delivered with an average click-through rate of 0.21%., Comment: Published at the 41st European Conference on Information Retrieval (ECIR) 2019

The main contribution of this paper is to introduce and describe a new recommender-systems dataset (RARD II). It is based on data from Mr. DLib, a recommender-system as-a-service in the digital library and reference-management-software domain. As such, RARD II complements datasets from other domains such as books, movies, and music. The dataset encompasses 94m recommendations, delivered in the two years from September 2016 to September 2018. The dataset covers an item-space of 24m unique items. RARD II provides a range of rich recommendation data, beyond conventional ratings. For example, in addition to the usual (implicit) ratings matrices, RARD II includes the original recommendation logs, which provide a unique insight into many aspects of the algorithms that generated the recommendations. The logs enable researchers to conduct various analyses about a real-world recommender system. This includes the evaluation of meta-learning approaches for predicting algorithm performance. In this paper, we summarise the key features of this dataset release, describe how it was generated and discuss some of its unique features. Compared to its predecessor RARD, RARD II contains 64% more recommendations, 187% more features (algorithms, parameters, and statistics), 50% more clicks, 140% more documents, and one additional service partner (JabRef).

The effectiveness of recommendation algorithms is typically assessed with evaluation metrics such as root mean square error, F1, or click through rates, calculated over entire datasets. The best algorithm is typically chosen based on these overall metrics. However, there is no single-best algorithm for all users, items, and contexts. Choosing a single algorithm based on overall evaluation results is not optimal. In this paper, we propose a meta-learning-based approach to recommendation, which aims to select the best algorithm for each user-item pair. We evaluate our approach using the MovieLens 100K and 1M datasets. Our approach (RMSE, 100K: 0.973; 1M: 0.908) did not outperform the single-best algorithm, SVD++ (RMSE, 100K: 0.942; 1M: 0.887). We also develop a distinction between meta-learners that operate per-instance (micro-level), per-data subset (mid-level), and per-dataset (global level). Our evaluation shows that a hypothetically perfect micro-level meta-learner would improve RMSE by 25.5% for the MovieLens 100K and 1M datasets, compared to the overall-best algorithms used.

"Position bias" describes the tendency of users to interact with items on top of a list with higher probability than with items at a lower position in the list, regardless of the items' actual relevance. In the domain of recommender systems, particularly recommender systems in digital libraries, position bias has received little attention. We conduct a study in a real-world recommender system that delivered ten million related-article recommendations to the users of the digital library Sowiport, and the reference manager JabRef. Recommendations were randomly chosen to be shuffled or non-shuffled, and we compared click-through rate (CTR) for each rank of the recommendations. According to our analysis, the CTR for the highest rank in the case of Sowiport is 53% higher than expected in a hypothetical non-biased situation (0.189% vs. 0.123%). Similarly, in the case of Jabref the highest rank received a CTR of 1.276%, which is 87% higher than expected (0.683%). A chi-squared test confirms the strong relationship between the rank of the recommendation shown to the user and whether the user decided to click it (p < 0.01 for both Jabref and Sowiport). Our study confirms the findings from other domains, that recommendations in the top positions are more often clicked, regardless of their actual relevance.

Creating scientific publications is a complex process, typically composed of a number of different activities, such as designing the experiments, data preparation, programming software and writing and editing the manuscript. The information about the contributions of individual authors of a paper is important in the context of assessing authors' scientific achievements. Some publications in biomedical disciplines contain a description of authors' roles in the form of a short section written in natural language, typically entitled "Authors' contributions". In this paper, we present an analysis of roles commonly appearing in the content of these sections, and propose an algorithm for automatic extraction of authors' roles from natural language text in scientific publications. During the first part of the study, we used clustering techniques, as well as Open Information Extraction (OpenIE), to semi-automatically discover the most popular roles within a corpus of 2,000 contributions sections obtained from PubMed Central resources. The roles discovered by our approach include: experimenting (1,743 instances, 17% of the entire role set within the corpus), analysis (1,343, 16%), study design (1,132, 13%), interpretation (879, 10%), conceptualization (865, 10%), paper reading (823, 10%), paper writing (724, 8%), paper review (501, 6%), paper drafting (351, 4%), coordination (319, 4%), data collection (76, 1%), paper review (41, 0.5%) and literature review (41, 0.5%). Discovered roles were then used to automatically build a training set for the supervised role extractor, based on Naive Bayes algorithm. According to the evaluation we performed, the proposed role extraction algorithm is able to extract the roles from the text with precision 0.71, recall 0.49 and F1 0.58.

Bibliographic reference parsing refers to extracting machine-readable metadata, such as the names of the authors, the title, or journal name, from bibliographic reference strings. Many approaches to this problem have been proposed so far, including regular expressions, knowledge bases and supervised machine learning. Many open source reference parsers based on various algorithms are also available. In this paper, we apply, evaluate and compare ten reference parsing tools in a specific business use case. The tools are Anystyle-Parser, Biblio, CERMINE, Citation, Citation-Parser, GROBID, ParsCit, PDFSSA4MET, Reference Tagger and Science Parse, and we compare them in both their out-of-the-box versions and versions tuned to the project-specific data. According to our evaluation, the best performing out-of-the-box tool is GROBID (F1 0.89), followed by CERMINE (F1 0.83) and ParsCit (F1 0.75). We also found that even though machine learning-based tools and tools based on rules or regular expressions achieve on average similar precision (0.77 for ML-based tools vs. 0.76 for non-ML-based tools), applying machine learning-based tools results in a recall three times higher than in the case of non-ML-based tools (0.66 vs. 0.22). Our study also confirms that tuning the models to the task-specific data results in the increase in the quality. The retrained versions of reference parsers are in all cases better than their out-of-the-box counterparts; for GROBID F1 increased by 3% (0.92 vs. 0.89), for CERMINE by 11% (0.92 vs. 0.83), and for ParsCit by 16% (0.87 vs. 0.75)., Comment: to appear in Proceedings of Joint Conference on Digital Libraries 2018

Coronene, a polyaromatic hydrocarbon, has been crystallized for the first time in a different polymorph using a crystal growth method that utilizes magnetic fields to access a unit cell configuration that was hitherto unknown. Crystals grown in magnetic field of 1 T are larger, have a different appearance to those grown in zero field and retain their structure in ambient conditions. We identify the new form, beta-coronene, as the most stable at low temperatures. As a result of the new supramolecular configuration we report significantly altered electronic, optical and mechanical properties., Comment: 32 pages, 17 figures

The idea of implicit representation of graphs was introduced in [S. Kannan, M. Naor, S. Rudich, Implicit representation of graphs, SIAM J. Discrete Mathematics, 5 (1992) 596--603] and can be defined as follows. A representation of an $n$-vertex graph $G$ is said to be implicit if it assigns to each vertex of $G$ a binary code of length $O(\log n)$ so that the adjacency of two vertices is a function of their codes. Since an implicit representation of an $n$-vertex graph uses $O(n\log n)$ bits, any class of graphs admitting such a representation contains $2^{O(n\log n)}$ labelled graphs with $n$ vertices. In the terminology of [J. Balogh, B. Bollob\'{a}s, D. Weinreich, The speed of hereditary properties of graphs, J. Combin. Theory B 79 (2000) 131--156] such classes have at most factorial speed of growth. In this terminology, the implicit graph conjecture can be stated as follows: every class with at most factorial speed of growth which is hereditary admits an implicit representation. The question of deciding whether a given hereditary class has at most factorial speed of growth is far from being trivial. In the present paper, we introduce a number of tools simplifying this question. Some of them can be used to obtain a stronger conclusion on the existence of an implicit representation. We apply our tools to reveal new hereditary classes with the factorial speed of growth. For many of them we show the existence of an implicit representation.

Many common diseases have a complex genetic basis in which large numbers of genetic variations combine with environmental and lifestyle factors to determine risk. However, quantifying such polygenic effects and their relationship to disease risk has been challenging. In order to address these difficulties we developed a global measure of the information content of an individual's genome relative to a reference population, which may be used to assess differences in global genome structure between cases and appropriate controls. Informally this measure, which we call relative genome information (RGI), quantifies the relative "disorder" of an individual's genome. In order to test its ability to predict disease risk we used RGI to compare single nucleotide polymorphism genotypes from two independent samples of women with early-onset breast cancer with three independent sets of controls. We found that RGI was significantly elevated in both sets of breast cancer cases in comparison with all three sets of controls, with disease risk rising sharply with RGI (odds ratio greater than 12 for the highest percentile RGI). Furthermore, we found that these differences are not due to associations with common variants at a small number of disease-associated loci, but rather are due to the combined associations of thousands of markers distributed throughout the genome. Our results indicate that the information content of an individual's genome may be used to measure the risk of a complex disease, and suggest that early-onset breast cancer has a strongly polygenic basis.

The paper [J. Balogh, B. Bollob\'{a}s, D. Weinreich, A jump to the Bell number for hereditary graph properties, J. Combin. Theory Ser. B 95 (2005) 29--48] identifies a jump in the speed of hereditary graph properties to the Bell number $B_n$ and provides a partial characterisation of the family of minimal classes whose speed is at least $B_n$. In the present paper, we give a complete characterisation of this family. Since this family is infinite, the decidability of the problem of determining if the speed of a hereditary property is above or below the Bell number is questionable. We answer this question positively by showing that there exists an algorithm which, given a finite set $\mathcal{F}$ of graphs, decides whether the speed of the class of graphs containing no induced subgraphs from the set $\mathcal{F}$ is above or below the Bell number. For properties defined by infinitely many minimal forbidden induced subgraphs, the speed is known to be above the Bell number., Comment: 19 pages; v2, v3: minor improvements in exposition