Shamsuddeen Muhammad, Idris Abdulmumin, Abinew Ayele, Nedjma Ousidhoum, David Adelani, Seid Yimam, Ibrahim Ahmad, Meriem Beloucif, Saif Mohammad, Sebastian Ruder, Oumaima Hourrane, Alipio Jorge, Pavel Brazdil, Felermino Ali, Davis David, Salomey Osei, Bello Shehu-Bello, Falalu Lawan, Tajuddeen Gwadabe, Samuel Rutunda, Tadesse Belay, Wendimu Messelle, Hailu Balcha, Sisay Chala, Hagos Gebremichael, Bernard Opoku, Stephen Arthur. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 2023.

Visualizing and examining the intellectual landscape and evolution of scientific communities to support collaboration is crucial for multiple research purposes. In some cases, measuring similarities and matching patterns between research publication document sets can help to identify people with similar interests for building research collaboration networks and university–industry linkages. The premise of this work is assessing feasibility for resolving ambiguous cases in similarity detection to determine authorship with natural language processing (NLP) techniques so that crowdsourcing is applied only in instances that require human judgment. Using an NLP-crowdsourcing convergence strategy, we can reduce the costs of microtask crowdsourcing while saving time and maintaining disambiguation accuracy over large datasets. This article contributes a next-gen crowd-artificial intelligence framework that used an ensemble of term frequency-inverse document frequency and bidirectional encoder representation from transformers to obtain similarity rankings for pairs of scientific documents. A sequence of content-based similarity tasks was created using a crowd-powered interface for solving disambiguation problems. Our experimental results suggest that an adaptive NLP-crowdsourcing hybrid framework has advantages for inter-researcher similarity detection tasks where fully automatic algorithms provide unsatisfactory results, with the goal of helping researchers discover potential collaborators using data-driven approaches.

The Traveling Salesman Problem (TSP) is one of the most studied optimization problems.Various metaheuristics (MHs) have been proposed and investigated on many instances of this problem.It is widely accepted that the best MH varies for different instances.Ideally, one should be able to recommend the best MHs for a new TSP instance without having to execute them.However, this is a very difficult task.We address this task by using a meta-learning approach based on label ranking algorithms.These algorithms build a mapping that relates the characteristics of those instances (i.e., the meta-features) with the relative performance (i.e., the ranking) of MHs, based on (meta-)data extracted from TSP instances that have been already solved by those MHs.The success of this approach depends on the quality of the meta-features that describe the instances.In this work, we investigate four different sets of meta-features based on different measurements of the properties of TSP instances: edge and vertex measures, complex network measures, properties from the MHs, and subsampling landmarkers properties.The models are investigated in four different TSP scenarios presenting symmetry and connection strength variations.The experimental results indicate that meta-learning models can accurately predict rankings of MHs for different TSP scenarios.Good solutions for the investigated TSP instances can be obtained from the prediction of rankings of MHs, regardless of the learning algorithm used at the metalevel.The experimental results also show that the definition of the set of meta-features has an important impact on the quality of the solutions obtained.

Data Mining (DM) researchers often focus on the development and testing of models for a single decision (e.g., direct mailing, churn detection, etc.). In practice, however, multiple decisions have often to be made simultaneously which are not independent and the best global solution is often not th e combination of the best individual solutions. This problem can be addressed by searching for the overall best solution by using optimization methods based on the predictions made by the DM models. We describe one case study were this approach was used to optimize the layout of a retail store in order to maximize predicted sales. A metaheuristic is used to search different hypothesis of space allocations for multiple product categories, guided by the predictions made by regression models that estimate the sales for each category based on the assigned space. We test three metaheuristics and three regression algorithms on this task. Results show that the Particle Swam Optimization method guided by the models obtained with Random Forests and Support Vector Machines models obtain good results. We also provide insights about the relationship between the correctness of the regression models and the metaheuristics performance.

Several algorithms based on PageRank algorithm have been proposed to rank the document sentences in the multi-document summarization field and LexRank and T-LexRank algorithms are well known examples. In literature different concepts such as weighted inter-cluster edge, cluster-sensitive graph mode l and document-sensitive graph model have been proposed to improve LexRank and T-LexRank algorithms (e.g. DsR-G, DsR-Q) for multi-document summarization. In this paper, a density-based graph model for multi-document summarization is proposed by adding the concept of density to LexRank and T-LexRank algorithms. The resulting generic multi-document summarization systems, DensGS and DensGSD were evaluated on DUC 2004 while the query-based variants, DensQS, DensQSD were evaluated on DUC 2006, DUC 2007 and TAC 2010 task A. ROUGE measure was used in the evaluation. Experimental results show that density concept improves LexRank and T-LexRank algorithms and outperforms previous graph-based models (DsR-G and DsR-Q) in generic and query-based multi-document summarization tasks. Furthermore, the comparison of the number of iterations indicates that the density-based algorithm is faster than the other algorithms based on PageRank.

We propose a Multi-Agent framework to analyze the dynamics of organizational survival in cooperation networks. Firms can decide to cooperate horizontally (in the same market) or vertically with otherfirms that belong to the supply chain. Cooperation decisions are based on economic variables, We have defined a variant of the density dependence model to set up the dynamics of the survival in the simulation. To validate our model, we have used empirical outputs obtained in previous studies from the automobile manufacturing sector. We have observed that firms and networks proliferate in the regions with lower marginal costs, but new networks keep appearing and disappearing in regions with higher marginal costs.

This paper is concerned with the problem of predicting relative performance of classification algorithms. It focusses on methods that use results on small samples and discusses the shortcomings of previous approaches. A new variant is proposed that exploits, as some previous approaches, metalearning. The method requires that experiments be conducted on few samples. The information gathered is used to identify the nearest learning curve for which the sampling procedure was carried out fully. This in turn permits to generate a prediction regards the relative performance of algorithms. Experimental evaluation shows that the method competes well with previous approaches and provides quite good and practical solution to this problem. 1.

The aim of behavioural cloning is to synthesize artificial controllers that are robust and comprehensible to human understanding. To attain the two objectives we propose the use of the Incremental Correction model that is based on a closed-loop control strategy to model the reactive aspects of human control skills. We have investigated the use of three different representations to encode the artificial controllers: univariate decision trees as induced by C4.5; multivariate decision and regression trees as induced by cart and; clausal theories induced by an Inductive Logic Programming (ILP) system. We obtained an increase in robustness and a lower complexity of the controllers when compared with results using other models. The controllers synthesized by cart revealed to be the most robust. The ILP system produced the simpler encodings.

In this paper we present system Ltree for propositional supervised learning. Ltree is able to define decision surfaces both orthogonal and oblique to the axes defined by the attributes of the input space. This is done combining a decision tree with a linear discriminant by means of constructive induction. At each decision node Ltree defines a new instance space by insertion of new attributes that are projections of the examples that fall at this node over the hyper-planes given by a linear discriminant function. This new instance space is propagated down through the tree. Tests based on those new attributes are oblique with respect to the original input space. Ltree is a probabilistic tree in the sense that it outputs a class probability distribution for each query example. The class probability distribution is computed at learning time, taking into account the different class distributions on the path from the root to the actual node. We have carried out experiments on twenty one benchmark datasets and compared our system with other well known decision tree systems (orthogonal and oblique) like C4.5, OC1, LMDT, and CART. On these datasets we have observed that our system has advantages in what concerns accuracy and learning times at statistically significant confidence levels.