Deep context of citations using machine-learning models in scholarly full-text articles

Abstract

Information retrieval systems for scholarly literature rely heavily not only on text matching but on semantic-and context-based features. Readers nowadays are deeply interested in how important an article is, its purpose and how influential it is in follow-up research work. Numerous techniques to tap the power of machine learning and artificial intelligence have been developed to enhance retrieval of the most influential scientific literature. In this paper, we compare and improve on four existing state-of-the-art techniques designed to identify influential citations. We consider 450 citations from the Association for Computational Linguistics corpus, classified by experts as either important or unimportant, and further extract 64 features based on the methodology of four state-of-the-art techniques. We apply the Extra-Trees classifier to select 29 best features and apply the Random Forest and Support Vector Machine classifiers to all selected techniques. Using the Random Forest classifier, our supervised model improves on the state-of-the-art method by 11.25%, with 89% Precision-Recall area under the curve. Finally, we present our deep-learning model, the Long Short-Term Memory network, that uses all 64 features to distinguish important and unimportant citations with 92.57% accuracy.

Citation Counts as a Measure for Scientific Impact

“…All in all, the concept of Scite can bring much value towards the scientific world, a tool that categorizes citations in a basic way distinguishing between mentioning, contradicting and supporting can help a scientist better interpret citation counts. There are other deep learning models (Hassan, Imran, Iqbal, Aljohani & Nawaz, 2018) which analyze citations using language processing reaching a similar accuracy (91% is the best result stated in the article) as the Scite model but those models only give a level of importance, which is less valuable than a classification. Based on our data Scite reaches an accuracy of 92% percent.…”

“…Based on our data Scite reaches an accuracy of 92% percent. The models mentioned in Hassan et al (2018) measure something different than the model used by Scite, therefore relating the accuracy percentages is illogical. When Scite is able to match the number of articles held within their database to WoS or Scopus, and increase accuracy it has the capability to outvalue other sources of citations.…”

Enhancement of Deep Learning in Image Classification Performance Using Xception with the Swish Activation Function for Colorectal Polyp Preliminary Screening

“…Regarding training and validation history, 50 epochs seemed good enough for our experiment [61,62], as illustrated in Figure 7. Both the original Xception and Xception with Swish raised training accuracy up to 100%, as shown in Figure 7a, and decreased training and validation loss very low, down to 0.0001% and 0.0002%, respectively, as shown in Figure 7b.…”

Linking Work-Family Conflict (WFC) and Talent Management: Insights from a Developing Country

“…Indeed, this is the scope of research our team embarked on already. In this vein, we plan to extend this work by, on the one hand incorporating more textual and non-textual data [70][71][72][73][74] and, on the other hand, by applying the findings in diverse contexts [75]. Apart from offering direct insights into respective policy considerations, these might then also be the useful in context of natural language processing models [76][77][78][79][80] and optimization techniques [81,82].…”