publications

2024

1. 

   Causally estimating the effect of YouTube’s recommender system using counterfactual bots

   Homa Hosseinmardi, Amir Ghasemian, Miguel Rivera-Lanas, and 3 more authors

   Proceedings of the National Academy of Sciences, 2024

   Abs Bib HTML

   In recent years, critics of online platforms have raised concerns about the ability of recommendation algorithms to amplify problematic content, with potentially radicalizing consequences. However, attempts to evaluate the effect of recommenders have suffered from a lack of appropriate counterfactuals – what a user would have viewed in the absence of algorithmic recommendations – and hence cannot disentangle the effects of the algorithm from a user’s intentions. Here we propose a method that we call "counterfactual bots" to causally estimate the role of algorithmic recommendations on the consumption of highly partisan content. By comparing bots that replicate real users’ consumption patterns with "counterfactual" bots that follow rule-based trajectories, we show that, on average, relying exclusively on the recommender results in less partisan consumption, where the effect is most pronounced for heavy partisan consumers. Following a similar method, we also show that if partisan consumers switch to moderate content, YouTube’s sidebar recommender "forgets" their partisan preference within roughly 30 videos regardless of their prior history, while homepage recommendations shift more gradually towards moderate content. Overall, our findings indicate that, at least on YouTube, individual consumption patterns mostly reflect individual preferences, where algorithmic recommendations play, if anything, a moderating role.

   @article{hosseinmardi2023causally,
     title = {Causally estimating the effect of YouTube's recommender system using counterfactual bots},
     author = {Hosseinmardi, Homa and Ghasemian, Amir and Rivera-Lanas, Miguel and Ribeiro, Manoel Horta and West, Robert and Watts, Duncan J},
     journal = {Proceedings of the National Academy of Sciences},
     volume = {121},
     number = {8},
     pages = {e2313377121},
     year = {2024},
     publisher = {National Acad Sciences},
   }

2. 

   Sequential Stacking Link Prediction Algorithms for Temporal Networks

   Xie He, Amir Ghasemian, Eun Lee, and 2 more authors

   Nature Communications, 2024

   Abs Bib HTML Code

   Link prediction algorithms are indispensable tools in many scientific applications by speeding up network data collection and imputing missing connections. However, in many systems, links change over time and it remains unclear how to optimally exploit such temporal information for link predictions in such networks. Here, we show that many temporal topological features, in addition to having high computational cost, are less accurate in temporal link prediction than sequentially stacked static network features. This sequential stacking link prediction method uses 41 static network features that avoids detailed feature engineering choices and is capable of learning a highly accurate predictive distribution of future connections from historical data. We demonstrate that this algorithm works well for both partially observed and completely unobserved target layers, and achieves near-optimal AUC on two temporal stochastic block models. Finally, we empirically illustrate that stacking multiple predictive methods together further improves performance on 19 real-world temporal networks from different domains.

   @article{he2023sequential,
     title = {Sequential Stacking Link Prediction Algorithms for Temporal Networks},
     author = {He, Xie and Ghasemian, Amir and Lee, Eun and Clauset, Aaron and Mucha, Peter},
     journal = {Nature Communications},
     volume = {15},
     issue = {1},
     pages = {1364},
     year = {2024},
   }

3. 

   Link Prediction Accuracy on Real-World Networks Under Non-Uniform Missing Edge Patterns

   Xie He, Amir Ghasemian, Eun Lee, and 3 more authors

   arXiv preprint arXiv:2401.15140, 2024

   Abs Bib HTML

   Real-world network datasets are typically obtained in ways that fail to capture all links, and there are many different non-uniform ways in which real data might be missing. Nevertheless, uniform missing data is a common assumption made when no additional information is available about the underlying ”missingness function.” To investigate the impact of different missingness patterns on link prediction accuracy, we employ 9 link prediction algorithms from 4 different families to analyze 20 different missingness functions categorized into 5 groups. By studying 250 real-world network datasets, we illustrate that different prediction algorithms exhibit significant differences in accuracy contingent upon both the dataset domain and the nature of the missingness pattern. Our study thereby provides guidance for selecting appropriate prediction algorithms when encountering diverse patterns of missing data across various domains, emphasizing the importance of considering the specific characteristics of the dataset for effective link prediction.

   @article{he2023sampling,
     title = {Link Prediction Accuracy on Real-World Networks Under Non-Uniform Missing Edge Patterns},
     author = {He, Xie and Ghasemian, Amir and Lee, Eun and Schwarze, Alice and Clauset, Aaron and Mucha, Peter J},
     journal = {arXiv preprint arXiv:2401.15140},
     year = {2024},
   }

2023

1. 

   The Enmity Paradox

   Amir Ghasemian, and Nicholas A Christakis

   Scientific Reports, 2023

   Abs Bib HTML Code

   The "friendship paradox" of social networks states that, on average, "your friends have more friends than you do." Here, we theoretically and empirically explore a related and overlooked paradox we refer to as the "enmity paradox." We use empirical data from 24,687 people living in 176 villages in rural Honduras. We show that, for a real negative undirected network (created by symmetrizing antagonistic interactions), the paradox exists as it does in the positive world. Specifically, a person’s enemies have more enemies, on average, than a person does. Furthermore, in a mixed world of positive and negative ties, we study the conditions for the existence of the paradox, both theoretically and empirically, finding that, for instance, a person’s friends typically have more enemies than a person does. We also confirm the "generalized" enmity paradox for nontopological attributes in real data, analogous to the generalized friendship paradox (e.g., the claim that a person’s enemies are richer, on average, than a person is). As a consequence, the naturally occurring variance in the degree distribution of both friendship and antagonism in social networks can skew people’s perceptions of the social world.

   @article{ghasemian2023enmity,
     title = {The Enmity Paradox},
     author = {Ghasemian, Amir and Christakis, Nicholas A},
     journal = {Scientific Reports},
     volume = {13},
     issue = {1},
     pages = {20040},
     year = {2023},
   }

2. 

   Subspace based DOA estimation of DS-CDMA signals

   Amir Ghasemian, Ali Olfat, and Mojtaba Amiri

   Telecommunication Systems, 2023

   Abs Bib HTML

   This paper presents a subspace blind method to estimate the direction of arrival of direct sequence code division multiple access signals in a multipath fading environment. The proposed method is based on signal/noise subspace approach. For enhancing the results, the problem is formulated based on both signal and noise subspaces to exploit structures of desired signal, self interference and multiple access interference simultaneously. The main idea in this paper is based on fitting the extended subspace spanned by the desired signature waveform in all different paths into the estimated extended signal subspace. The proposed method is blind in the sense that it utilizes only the desired user’s code and its corresponding path delays. The performance of the proposed method is compared to the Cramer–Rao lower bound. We also propose a method for estimating relative power of different paths in multipath code division multiple access signals.

   @article{ghasemian2023subspace,
     title = {Subspace based DOA estimation of DS-CDMA signals},
     author = {Ghasemian, Amir and Olfat, Ali and Amiri, Mojtaba},
     journal = {Telecommunication Systems},
     pages = {1--12},
     year = {2023},
     publisher = {Springer},
   }

3. 

   Tensor embedding: a supervised framework for human behavioral data mining and prediction

   Homa Hosseinmardi, Amir Ghasemian, Shrikanth Narayanan, and 2 more authors

   In 2023 IEEE 11th International Conference on Healthcare Informatics (ICHI), 2023

   Abs Bib HTML

   Today’s densely instrumented world offers tremendous opportunities for continuous acquisition and analysis of multimodal sensor data providing temporal characterization of an individual’s behaviors. Is it possible to efficiently couple such rich sensor data with predictive modeling techniques to provide contextual, and insightful assessments of individual performance and wellbeing? Prediction of different aspects of human behavior from these noisy, incomplete, and heterogeneous bio-behavioral temporal data is a challenging problem, beyond unsupervised discovery of latent structures. We propose a Supervised Tensor Embedding (STE) algorithm for high dimension multimodal data with join decomposition of input and target variable. Furthermore, we show that features selection will help to reduce the contamination in the prediction and increase the performance. The efficiently of the methods was tested via two different real world datasets.

   @inproceedings{hosseinmardi2018tensor,
     title = {Tensor embedding: a supervised framework for human behavioral data mining and prediction},
     author = {Hosseinmardi, Homa and Ghasemian, Amir and Narayanan, Shrikanth and Lerman, Kristina and Ferrara, Emilio},
     booktitle = {2023 IEEE 11th International Conference on Healthcare Informatics (ICHI)},
     pages = {91--100},
     year = {2023},
     organization = {IEEE},
   }

2021

1. 

   Examining the consumption of radical content on YouTube

   Homa Hosseinmardi, Amir Ghasemian, Aaron Clauset, and 3 more authors

   Proceedings of the National Academy of Sciences, 2021

   Abs Bib HTML

   Although it is under-studied relative to other social media platforms, YouTube is arguably the largest and most engaging online media consumption platform in the world. Recently, YouTube’s scale has fueled concerns that YouTube users are being radicalized via a combination of biased recommendations and ostensibly apolitical “anti-woke” channels, both of which have been claimed to direct attention to radical political content. Here we test this hypothesis using a representative panel of more than 300,000 Americans and their individual-level browsing behavior, on and off YouTube, from January 2016 through December 2019. Using a labeled set of political news channels, we find that news consumption on YouTube is dominated by mainstream and largely centrist sources. Consumers of far-right content, while more engaged than average, represent a small and stable percentage of news consumers. However, consumption of “anti-woke” content, defined in terms of its opposition to progressive intellectual and political agendas, grew steadily in popularity and is correlated with consumption of far-right content off-platform. We find no evidence that engagement with far-right content is caused by YouTube recommendations systematically, nor do we find clear evidence that anti-woke channels serve as a gateway to the far right. Rather, consumption of political content on YouTube appears to reflect individual preferences that extend across the web as a whole.

   @article{hosseinmardi2021examining,
     title = {Examining the consumption of radical content on YouTube},
     author = {Hosseinmardi, Homa and Ghasemian, Amir and Clauset, Aaron and Mobius, Markus and Rothschild, David M and Watts, Duncan J},
     journal = {Proceedings of the National Academy of Sciences},
     volume = {118},
     number = {32},
     pages = {e2101967118},
     year = {2021},
     publisher = {National Acad Sciences},
   }

2020

1. 

   Stacking models for nearly optimal link prediction in complex networks

   Amir Ghasemian, Homa Hosseinmardi, Aram Galstyan, and 2 more authors

   Proceedings of the National Academy of Sciences, 2020

   Abs Bib HTML Code

   Most real-world networks are incompletely observed. Algorithms that can accurately predict which links are missing can dramatically speed up network data collection and improve network model validation. Many algorithms now exist for predicting missing links, given a partially observed network, but it has remained unknown whether a single best predictor exists, how link predictability varies across methods and networks from different domains, and how close to optimality current methods are. We answer these questions by systematically evaluating 203 individual link predictor algorithms, representing three popular families of methods, applied to a large corpus of 550 structurally diverse networks from six scientific domains. We first show that individual algorithms exhibit a broad diversity of prediction errors, such that no one predictor or family is best, or worst, across all realistic inputs. We then exploit this diversity using network-based metalearning to construct a series of “stacked” models that combine predictors into a single algorithm. Applied to a broad range of synthetic networks, for which we may analytically calculate optimal performance, these stacked models achieve optimal or nearly optimal levels of accuracy. Applied to real-world networks, stacked models are superior, but their accuracy varies strongly by domain, suggesting that link prediction may be fundamentally easier in social networks than in biological or technological networks. These results indicate that the state of the art for link prediction comes from combining individual algorithms, which can achieve nearly optimal predictions. We close with a brief discussion of limitations and opportunities for further improvements.

   @article{ghasemian2020stacking,
     title = {Stacking models for nearly optimal link prediction in complex networks},
     author = {Ghasemian, Amir and Hosseinmardi, Homa and Galstyan, Aram and Airoldi, Edoardo M and Clauset, Aaron},
     journal = {Proceedings of the National Academy of Sciences},
     volume = {117},
     number = {38},
     pages = {23393--23400},
     year = {2020},
     publisher = {National Acad Sciences},
   }

2019

1. 

   Limits of model selection, link prediction, and community detection

   University of Colorado at Boulder, 2019

   Bib HTML

   @phdthesis{ghasemian2019limits,
     title = {Limits of model selection, link prediction, and community detection},
     year = {2019},
     school = {University of Colorado at Boulder},
   }

2. 

   Evaluating overfit and underfit in models of network community structure

   Amir Ghasemian, Homa Hosseinmardi, and Aaron Clauset

   IEEE Transactions on Knowledge and Data Engineering, 2019

   Abs Bib HTML Code

   A common graph mining task is community detection, which seeks an unsupervised decomposition of a network into groups based on statistical regularities in network connectivity. Although many such algorithms exist, community detection’s No Free Lunch theorem implies that no algorithm can be optimal across all inputs. However, little is known in practice about how different algorithms over or underfit to real networks, or how to reliably assess such behavior across algorithms. Here, we present a broad investigation of over and underfitting across 16 state-of-the-art community detection algorithms applied to a novel benchmark corpus of 572 structurally diverse real-world networks. We find that (i) algorithms vary widely in the number and composition of communities they find, given the same input; (ii) algorithms can be clustered into distinct high-level groups based on similarities of their outputs on real-world networks; (iii) algorithmic differences induce wide variation in accuracy on link-based learning tasks; and, (iv) no algorithm is always the best at such tasks across all inputs. Finally, we quantify each algorithm’s overall tendency to over or underfit to network data using a theoretically principled diagnostic, and discuss the implications for future advances in community detection.

   @article{ghasemian2019evaluating,
     title = {Evaluating overfit and underfit in models of network community structure},
     author = {Ghasemian, Amir and Hosseinmardi, Homa and Clauset, Aaron},
     journal = {IEEE Transactions on Knowledge and Data Engineering},
     volume = {32},
     number = {9},
     pages = {1722--1735},
     year = {2019},
     publisher = {IEEE},
     href = {https://journals.aps.org/prx/abstract/10.1103/PhysRevX.6.031005},
   }

2016

1. 

   Detectability thresholds and optimal algorithms for community structure in dynamic networks

   Amir Ghasemian, Pan Zhang, Aaron Clauset, and 2 more authors

   Physical Review X, 2016

   Abs Bib HTML Code

   The detection of communities within a dynamic network is a common means for obtaining a coarse-grained view of a complex system and for investigating its underlying processes. While a number of methods have been proposed in the machine learning and physics literature, we lack a theoretical analysis of their strengths and weaknesses, or of the ultimate limits on when communities can be detected. Here, we study the fundamental limits of detecting community structure in dynamic networks. Specifically, we analyze the limits of detectability for a dynamic stochastic block model where nodes change their community memberships over time, but where edges are generated independently at each time step. Using the cavity method, we derive a precise detectability threshold as a function of the rate of change and the strength of the communities. Below this sharp threshold, we claim that no efficient algorithm can identify the communities better than chance. We then give two algorithms that are optimal in the sense that they succeed all the way down to this threshold. The first uses belief propagation, which gives asymptotically optimal accuracy, and the second is a fast spectral clustering algorithm, based on linearizing the belief propagation equations. These results extend our understanding of the limits of community detection in an important direction, and introduce new mathematical tools for similar extensions to networks with other types of auxiliary information.

   @article{ghasemian2016detectability,
     title = {Detectability thresholds and optimal algorithms for community structure in dynamic networks},
     author = {Ghasemian, Amir and Zhang, Pan and Clauset, Aaron and Moore, Cristopher and Peel, Leto},
     journal = {Physical Review X},
     volume = {6},
     number = {3},
     pages = {031005},
     year = {2016},
     publisher = {APS},
   }

2014

1. 

   Towards understanding cyberbullying behavior in a semi-anonymous social network

   Homa Hosseinmardi, Amir Ghasemian, Richard Han, and 2 more authors

   In 2014 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2014), 2014

   Abs Bib HTML

   Cyberbullying has emerged as an important and growing social problem, wherein people use online social networks and mobile phones to bully victims with offensive text, images, audio and video on a 24/7 basis. This paper studies negative user behavior in the Ask.fm social network, a popular new site that has led to many cases of cyberbullying, some leading to suicidal behavior.We examine the occurrence of negative words in Ask.fm’s question+answer profiles along with the social network of “likes” of questions+answers. We also examine properties of users with “cutting” behavior in this social network.

   @inproceedings{hosseinmardi2014towards,
     title = {Towards understanding cyberbullying behavior in a semi-anonymous social network},
     author = {Hosseinmardi, Homa and Ghasemian, Amir and Han, Richard and Lv, Qin and Mishra, Shivakant},
     booktitle = {2014 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2014)},
     pages = {244--252},
     year = {2014},
     organization = {IEEE},
   }