An official website of the United States government
Citations of scientific papers and patents reveal the knowledge flow and usually serve as the metric for evaluating their novelty and impacts in the field. Citation Forecasting thus has various applications in the real world. Existing works on citation forecasting typically exploit the sequential properties of citation events, without exploring the citation network. In this paper, we propose to explore both the citation network and the related citation event sequences which provide valuable information for future citation forecasting. We propose a novel Citation Network and Event Sequence (CINES) Model to encode signals in the citation network and related citation event sequences into various types of embeddings for decoding to the arrivals of future citations. Moreover, we propose a temporal network attention and three alternative designs of bidirectional feature propagation to aggregate the retrospective and prospective aspects of publications in the citation network, coupled with the citation event sequence embeddings learned by a two-level attention mechanism for the citation forecasting. We evaluate our models and baselines on both a U.S. patent dataset and a DBLP dataset. Experimental results show that our models outperform the state-of-the-art methods, i.e., RMTPP, CYAN-RNN, Intensity-RNN, and PC-RNN, reducing the forecasting error by 37.76% - 75.32%.
more »
« less
Citation Forecasting with Multi-Context Attention-Aided Dependency Modeling
Forecasting citations of scientific patents and publications is a crucial task for understanding the evolution and development of technological domains and for foresight into emerging technologies. By construing citations as a time series, the task can be cast into the domain of temporal point processes. Most existing work on forecasting with temporal point processes, both conventional and neural network-based, only performs single-step forecasting. In citation forecasting, however, the more salient goal isn-step forecasting: predicting the arrival of the nextncitations. In this article, we propose Dynamic Multi-Context Attention Networks (DMA-Nets), a novel deep learning sequence-to-sequence (Seq2Seq) model with a novel hierarchical dynamic attention mechanism for long-term citation forecasting. Extensive experiments on two real-world datasets demonstrate that the proposed model learns better representations of conditional dependencies over historical sequences compared to state-of-the-art counterparts and thus achieves significant performance for citation predictions.
more »
« less
- Award ID(s):
- 2153369
- PAR ID:
- 10505933
- Publisher / Repository:
- ACM
- Date Published:
- Journal Name:
- ACM Transactions on Knowledge Discovery from Data
- Volume:
- 18
- Issue:
- 6
- ISSN:
- 1556-4681
- Page Range / eLocation ID:
- 1 to 23
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Accurate prediction of scientific impact is important for scientists, academic recommender systems, and granting organizations alike. Existing approaches rely on many years of leading citation values to predict a scientific paper’s citations (a proxy for impact), even though most papers make their largest contributions in the first few years after they are published. In this paper, we tackle a new problem: predicting a new paper’s citation time series from the date of publication (i.e., without leading values). We propose HINTS, a novel end-to-end deep learning framework that converts citation signals from dynamic heterogeneous information networks (DHIN) into citation time series. HINTS imputes pseudo-leading values for a paper in the years before it is published from DHIN embeddings, and then transforms these embeddings into the parameters of a formal model that can predict citation counts immediately after publication. Empirical analysis on two real-world datasets from Computer Science and Physics show that HINTS is competitive with baseline citation prediction models. While we focus on citations, our approach generalizes to other “cold start” time series prediction tasks where relational data is available and accurate prediction in early timestamps is crucial.more » « less
-
Large quantities of asynchronous event sequence data such as crime records, emergence call logs, and financial transactions are becoming increasingly available from various fields. These event sequences often exhibit both long-term and short-term temporal dependencies. Variations of neural network based temporal point processes have been widely used for modeling such asynchronous event sequences. However, many current architectures including attention based point processes struggle with long event sequences due to computational inefficiency. To tackle the challenge, we propose an efficient sparse transformer Hawkes process (STHP), which has two components. For the first component, a transformer with a novel temporal sparse self-attention mechanism is applied to event sequences with arbitrary intervals, mainly focusing on short-term dependencies. For the second component, a transformer is applied to the time series of aggregated event counts, primarily targeting the extraction of long-term periodic dependencies. Both components complement each other and are fused together to model the conditional intensity function of a point process for future event forecasting. Experiments on real-world datasets show that the proposed STHP outperforms baselines and achieves significant improvement in computational efficiency without sacrificing prediction performance for long sequences.more » « less
-
Traditional citation analysis methods have been criticized because their theoretical base of statistical counts does not reflect the motive or judgment of citing authors. In particular, self-citations may give undue credits to a cited article or mislead scientific development. This research aims to answer the question of whether self-citation is biased by probing into the motives and context of citations. It takes an integrated and fine-grained view of self-citations by examining them via multiple lenses—polarity, density, and location of citations. In addition, it explores potential moderating effects of citation level and associations among location contexts of citations to the same references for the first time. We analyzed academic publications across different topics and disciplines using both qualitative and quantitative methods. The results provide evidence that self-citations are free of bias in terms of citation density and polarity uncertainty, but they can be biased with respect to positivity and negativity of citations. Furthermore, this study reveals impacts of self-citing behavior on some citation patterns involving citation density, location concentration, and associations. The examination of self-citing behavior from those new perspectives shed new lights on the nature and function of self-citing behavior.more » « less
-
Traffic data forecasting has become an integral part of the intelligent traffic system. Great efforts are spent developing tools and techniques to estimate traffic flow patterns. Many existing approaches lack the ability to model the complex and dynamic spatio-temporal relations in the traffic data, which are crucial in capturing the traffic dynamic. In this work, we propose AJSTGL, a novel adaptive joint spatio-temporal graph learning network for traffic data forecasting. The proposed model utilizes static and adaptive graph learning modules to capture the static and dynamic spatial traffic patterns and optimize the graph learning process. A sequence-to-sequence fusion model is proposed to learn the temporal correlation and combine the output of multiple parallelized encoders. We also develop a spatio-temporal graph transformer module to complement the sequence-to-sequence fusion module by dynamically capturing the time-evolving node relations in long-term intervals. Experiments on three large-scale traffic flow datasets demonstrate that our model could outperform other state-of-the-art baseline methods.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1145/3649140
