Abstract
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.

Abstract
Deep learning approaches have become popular in many different areas, including sentiment analysis (SA), because of their competitive performance. However, the downside of this approach is that they do not provide understandable explanations on how the sentiment values are calculated. In contrast, previous approaches that used sentiment lexicons can do that, but their performance is normally not high. To leverage the strengths of both approaches, we present a neuro-symbolic approach that combines deep learning (DL) and symbolic methods for SA tasks. The DL approach uses a pre-trained language model (PLM) to construct sentiment lexicon. The symbolic approach exploits the constructed sentiment lexicon and manually constructed shifter patterns to determine the sentiment of a sentence. Our experimental results show that the proposed approach leads to promising results with the additional advantage that sentiment predictions can be accompanied by understandable explanations.

Abstract
Africa is home to over 2,000 languages from more than six language families and has the highest linguistic diversity among all continents. These include 75 languages with at least one million speakers each. Yet, there is little NLP research conducted on African languages. Crucial to enabling such research is the availability of high-quality annotated datasets. In this paper, we introduce AfriSenti, a sentiment analysis benchmark that contains a total of >110,000 tweets in 14 African languages (Amharic, Algerian Arabic, Hausa, Igbo, Kinyarwanda, Moroccan Arabic, Mozambican Portuguese, Nigerian Pidgin, Oromo, Swahili, Tigrinya, Twi, Xitsonga, and Yorùbá) from four language families. The tweets were annotated by native speakers and used in the AfriSenti-SemEval shared task 1. We describe the data collection methodology, annotation process, and the challenges we dealt with when curating each dataset. We further report baseline experiments conducted on the different datasets and discuss their usefulness. ©2023 Association for Computational Linguistics.

Abstract

Abstract