The DOI matches its Crossref record; 60 of the 61 references it cites were found in a registry or an index.
DeepMood: Modeling Mobile Phone Typing Dynamics for Mood Detection
Open accessAbstract
The increasing use of electronic forms of communication presents new opportunities in the study of mental health, including the ability to investigate the manifestations of psychiatric diseases unobtrusively and in the setting of patients' daily lives. A pilot study to explore the possible connections between bipolar affective disorder and mobile phone usage was conducted. In this study, participants were provided a mobile phone to use as their primary phone. This phone was loaded with a custom keyboard that collected metadata consisting of keypress entry time and accelerometer movement. Individual character data with the exceptions of the backspace key and space bar were not collected due to privacy concerns. We propose an end-to-end deep architecture based on late fusion, named DeepMood, to model the multi-view metadata for the prediction of mood scores. Experimental results show that 90.31% prediction accuracy on the depression score can be achieved based on session-level mobile phone typing dynamics which is typically less than one minute. It demonstrates the feasibility of using mobile phone metadata to infer mood disturbance and severity.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Others focused on detecting depression by tracking medication, sleep patterns and actions [20], location [11] or keypress acceleration [165].”
From Human-centred artificial intelligence for mobile health sensing: challenges and opportunities (2023) · cites this paper as [165]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Diagnostic and Statistical Manual of Mental Disorders
American Psychiatric Association (2013) · doi:10.1176/appi.books.9780890425596 · Cited by 116,369
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 2×
“y, DeepMood provides a late fusion framework. It first models each view of the time series data separately using Gated Recurrent Unit (GRU) [12], a simplified version of Long Short-Term Memory (LSTM) [24]. It then fuses the output of the GRU from each view. As the GRU extracts a latent feature representation out of ea…”
MethodXGBoost: A Scalable Tree Boosting System
Chen, T., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939785 · Cited by 54,569 · cited 1×
“• XGB: The implementation of a tree boosting system from XGBoost6 [10] is used.”
MethodLearning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1179 · Cited by 25,545 · cited 11×
“A typical GRU is formulated as: rk = sigmoid(Wr xk + Ur hk−1) zk = sigmoid(Wzxk + Uzhk−1) h̃k = tanh(Wxk + U(rk ⊙ hk−1)) hk = zk ⊙ hk−1 + (1 − zk ) ⊙ h̃k (3) where ⊙ is the element-wise multiplication operator, a reset gate rk allows the GRU to forget the previously computed state hk−1, and an update gate zk balances…”
BackgroundMethodInfluentialKessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005) · Archives of General Psychiatry · doi:10.1001/archpsyc.62.6.593 · Cited by 20,391 · cited 1×
“There are 7 participants with bipolar I disorder that involves periods of severe mood episodes from mania to depression, 5 participants with bipolar II disorder which is a milder form of mood elevation, involving milder episodes of hypomania that alternate with periods of severe depression, and 8 participants with no…”
BackgroundNeural Machine Translation by Jointly Learning to Align and Translate
Bahdanau, D., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1409.0473 · Cited by 16,213 · cited 1×
“Inspired by the great success of deep RNNs in the applications of other sequence tasks, including speech recognition [20] and natural language processing [4, 34], in this work, we propose a deep architecture to model complex event sequences of mobile phone typing dynamics.”
Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1412.3555 · Cited by 11,008 · cited 1×
“The GRU has been attracting great attentions since it overcomes the vanishing gradient problem in traditional RNNs and is more efficient than the LSTM in some tasks [14].”
MethodTensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G. S., Davis, A., Dean, J., Devin, M., et al. (2016) · arXiv · doi:10.48550/arxiv.1603.04467 · Cited by 9,408
Speech recognition with deep recurrent neural networks
Graves, A., Mohamed, A.-R., & Hinton, G. (2013) · 2013 IEEE International Conference on Acoustics, Speech and Signal Processing · doi:10.1109/icassp.2013.6638947 · Cited by 8,893 · cited 1×
“Inspired by the great success of deep RNNs in the applications of other sequence tasks, including speech recognition [20] and natural language processing [4, 34], in this work, we propose a deep architecture to model complex event sequences of mobile phone typing dynamics.”
A Rating Scale for Mania: Reliability, Validity and Sensitivity
Young, R. C., Biggs, J. T., Ziegler, V. E., & Meyer, D. A. (1978) · British Journal of Psychiatry · doi:10.1192/bjp.133.5.429 · Cited by 8,744 · cited 7×
“Based on the Kolmogorov-Smirnov test, for 98.06% in HDRS and 99.52% in YMRS of the distribution pairs, we can reject the null hypothesis that two samples are drawn from the same distribution with significance level α = 0.01.”
MethodInfluential
Flagged first, then by citations
1 of the 61 references couldn’t be matched to a known paper.
Research works printed but not linked: 1
- Tijmen Tieleman and Geoffrey Hinton 2012 . Lecture 6.5-rmsprop: Divide the gradient by a running average of its recent magnitude . COURSERA: Neural Networks for Machine Learning , Vol. 4 , 2 (2012). Tijmen Tieleman and Geoffrey Hinton 2012. Lecture 6.5-rmsprop: Divide the gradient by a running average of its recent magnitude. COURSERA: Neural Networks for Machine Learning, Vol. 4, 2 (2012).
Cited by
Sorted by citations
Items 101–145 of 145
Keystroke Dynamics based Recognition Systems using Deep Learning: A SurveyPreprint
Tewari, A. (2022) · doi:10.36227/techrxiv.19532269 · Cited by 5
Gupta, A., Ritwik, & Chaurasiya, V. K. (2025) · IEEE Transactions on Green Communications and Networking · doi:10.1109/tgcn.2025.3624654 · Cited by 4
Unobtrusive inference of diurnal rhythms from smartphone data
Knol, L., Ross, M. K., Nagpal, A., Burns, A. P., Morrissey, Z. D., Hussain, F., Eisenlohr-Moul, T. A., Beckmann, C. F., Leow, A., & Marquand, A. F. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-02254-1 · Cited by 4
“Over the last eight years, we have developed and validated a leading smartphone-based platform, BiAffect, for unobtrusive monitoring of behavioural and mental health through keyboard typing dynamics and continuous sensor data 28 – 39 .”
Self-citationFrom Outside In: Profiling, Persuasion and Political Opinion in the Age of Big Data
Dhawan, S., & Hegelich, S. (2022) · Studies in neuroscience, psychology and behavioral economics · doi:10.1007/978-3-030-98546-2_10 · Cited by 6
Hari, S., Ajay, Sarcar, S., Sen, S., & Ghosh, S. (2022) · INTERNATIONAL CONFERENCE ON MULTIMODAL INTERACTION · doi:10.1145/3536221.3556603 · Cited by 4
Beyond non-inferior: how telepsychiatry technologies can lead to superior care
Zulueta, J., & Ajilore, O. (2020) · International Review of Psychiatry · doi:10.1080/09540261.2020.1835840 · Cited by 4
Self-citationStochastic modelling of transition dynamic of mixed mood episodes in bipolar disorder
Kumar, Y. K. A., & Saxena, A. K. (2021) · International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering · doi:10.11591/ijece.v12i1.pp620-629 · Cited by 4
Multimodal Privacy-preserving Mood Prediction from Mobile Data: A Preliminary StudyPreprint
Liu, T., Liang, P. P., Muszyński, M., Ishii, R., Brent, D. A., Auerbach, R. P., Allen, N. B., & Morency, L. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.02359 · Cited by 4
“In particular, they have found that typing patterns [2, 12, 15, 34], self-reporting apps [28], and wearable sensors [11, 13, 25] are particularly predictive for affect.”
BackgroundKeystroke Dynamics based Recognition Systems using Deep Learning: A SurveyPreprint
Tewari, A. (2022) · doi:10.36227/techrxiv.19532269.v1 · Cited by 4
A Privacy-Oriented Approach for Depression Signs Detection Based on Speech Analysis
Vitale, F., Carbonaro, B., Cordasco, G., Esposito, A. A., Marrone, S., Raimo, G., & Verde, L. (2021) · Electronics · doi:10.3390/electronics10232986 · Cited by 4
“To implement a system able to run on users’ mobile terminals, some authors propose the usage of typing pattern analysis [47].”
MethodPredicting Mood from Digital Footprints Using Frequent Sequential Context Patterns Features
Alibasa, M. J., Calvo, R. A., & Yacef, K. (2022) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2022.2073321 · Cited by 4
“Cao et al. (2017) attempted to predict mood from smartphone typing patterns, instead of digital usage sequence patterns.”
BackgroundPrivacy and Tracking in the Emerging Mobile Applications: A Survey
Boussada, R., Bouchaala, M., & Saı̈dane, L. A. (2023) · International Wireless Communications and Mobile Computing Conference (IWCMC) · doi:10.1109/iwcmc58020.2023.10182497 · Cited by 3
“Indeed, the authors in [16] investigated the possibility of establishing a relationship between bipolar affective disorder and mobile phone use.”
BackgroundInferring Mood-While-Eating with Smartphone Sensing and Community-Based Model Personalization
Bangamuarachchi, W., Chamantha, A., Meegahapola, L., Kim, H., Ruiz-Correa, S., Perera, I., & Gatica-Perez, D. (2023) · ACM Trans. Comput. Heal. · doi:10.1145/3723047 · Cited by 12
“Some of them includes physical movements [60] smartphone usage [25, 136], social contexts [31, 134], and sound [97].”
Chitale, V., Henry, J. D., Matthews, B., Cobham, V. E., & Baghaei, N. (2025) · JMIR Mental Health · doi:10.2196/70577 · Cited by 4
Sensor-Based Behavior Recognition
Yu, Z., & Wang, Z. (2020) · Springer eBooks · doi:10.1007/978-981-15-2109-6_3 · Cited by 3
Automatic role identification for research teams with ranking multi-view machines
Ni, W., Guo, H., Liu, T., & Zeng, Q. (2020) · Knowledge and Information Systems · doi:10.1007/s10115-020-01504-w · Cited by 2
Merrill, M. A., & Althoff, T. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2107.06097 · Cited by 4
“Behavioral data has been modeled and mined using deep learning techniques across a variety of domains, including human activity recognition (using CNN) [28], personalized fitness recommendation (using stacked LSTM [6]) [18], mood prediction (using RNN, GRU, or autoencoder) [2, 21, 22], stress prediction (using LSTM an…”
TYDR: Track Your Daily Routine
Beierle, F. (2021) · T-labs series in telecommunication services · doi:10.1007/978-3-030-68840-0_4 · Cited by 1
Multiscale Bidirectional Input Convolutional and Deep Neural Network for Human Activity Recognition
Qiu, Y., Lin, L., Yang, L., Li, D., Song, R., Xu, G., & Shen, S. (2021) · Wireless Communications and Mobile Computing · doi:10.1155/2021/7374177 · Cited by 1
AI in Healthcare: Malignant or Benign?
Lloyd, N., & Khuman, A. S. (2022) · Brain informatics and health · doi:10.1007/978-981-19-5272-2_1 · Cited by 1
A Privacy-Preserving Scheme For Convolutional Neural Network-Based Applications In Mobile Cloud
Xia, C., Hua, J., Tong, W., Xiong, Y., & Zhong, S. (2020) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme46284.2020.9102960 · Cited by 1
“Note that for each i,j,k ∈ B, we have i,j,k ∈ [0, 1] due to the structure of the neural network.”
BackgroundEdgeDIPN: a Unified Deep Intent Prediction Network Deployed at the Edge.
Guo, L. Z., Hua, L., Jia, R., Fang, F., Zhao, B., & Cui, B. (2020) · Very Large Data Bases · Cited by 1
Lan: Learning to Augment Noise Tolerance for Self-report Survey Labels
Lin, S., Faust, L. J., & Chawla, N. V. (2021) · IEEE International Conference on Pervasive Computing and Communications · doi:10.1109/percom50583.2021.9439109 · Cited by 1
“[3] and a sequence-to-sequence autoencoder to detect abnormal heart rates [41].”
Backgroundmotif2vec: Semantic-aware Representation Learning for Wearables' Time Series Data
Lin, S., Wu, X., & Chawla, N. V. (2021) · IEEE International Conference on Data Science and Advanced Analytics (DSAA) · doi:10.1109/dsaa53316.2021.9564120 · Cited by 1
Emotion detection from smartphone keyboard interactions
Mandi, S., Ghosh, S., De, P., & Mitra, B. (2022) · ACM Symposium on Applied Computing · doi:10.1145/3477314.3507159 · Cited by 1
“Recently, a significant volume of literature have explored the potential of the touch and keyboard interactions with the desktop keyboard or smartphone surfaces to infer user emotion, stress, and related affective states [1, 5, 7].”
BackgroundHussain, F., Stange, J. P., Langenecker, S. A., McInnis, M. G., Zulueta, J., Piscitello, A., Ross, M. K., Demos, A. P., Vesel, C., Rashidisabet, H., et al. (2022) · Studies in neuroscience, psychology and behavioral economics · doi:10.1007/978-3-030-98546-2_13 · Cited by 1
Self-citationConformal Learning for Quantifying Uncertainty in Psychiatric Condition PredictionsPreprint
Stewart, C., Millar, A., Facelli, J. C., & Gouripeddi, R. (2025) · Research Square · doi:10.21203/rs.3.rs-6458806/v1 · Cited by 1
Broad Learning for HealthcarePreprint
Cao, B. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1803.08978
“(This chapter was previously published as “DeepMood: Modeling Mobile Phone Typing Dynamics for Mood Detection (27)”, in Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), 2017, ACM.”
BackgroundInfluentialSelf-citationNeural Learning of Online Consumer Credit RiskPreprint
Wang, D., Wu, Q., & Zhang, W. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.01923
“Here, we employ a Multi-view Machines layer (Cao et al. 2017) for fusion.”
MethodMBead: Semi-supervised Multilabel Behaviour Anomaly Detection on Multivariate Temporal Sensory Data
Lin, S., Faust, L. J., D’Mello, S. K., Martinez, G. J., & Chawla, N. V. (2020) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata50022.2020.9378327 · Cited by 1
“For example, Cao et al . modeled the multi-view sensory data and predicted human mood with a end-to-end deep architecture [2].”
BackgroundMobile Sensing for Emotion Recognition
Piskioulis, O. (2021)
Progress on Computational Ethology
Ren, W., Yu, S. P., & Zhang, Y. (2021) · Chinese Science Bulletin (Chinese Version) · doi:10.1360/tb-2021-0213
Beierle, F. (2021) · T-labs series in telecommunication services · doi:10.1007/978-3-030-68840-0_3
Intelligent Behavioral Health Analysis in Youth Using Real-Time Smartphone and Wearable Data
Arora, A., Chakraborty, P., & Bhatia, M. P. S. (2022) · Algorithms for intelligent systems · doi:10.1007/978-981-19-3951-8_49
Bocu, R., & Bocu, D. (2023) · Lecture notes in networks and systems · doi:10.1007/978-3-031-28694-0_1
Priyadarshana, Y. H. P. P., Liang, Z., & Piumarta, I. (2023) · doi:10.2196/preprints.55181
Associations between keystroke and stylus metadata and depressive symptoms in adolescents (Preprint)Preprint
Jang, M., Cho, Y., Kim, D. H., Park, S., Park, S., Hur, J., Kim, M., Cho, K., Lee, C.-G., & Kwon, J. S. (2024) · doi:10.2196/preprints.56548
Sistema para el mapeo, control y monitoreo de síntomas depresivos en escolares
Cieza, A. E. G., & Monteza, G. C. (2023) · Universidad Peruana de Ciencias Aplicadas (UPC) · doi:10.19083/tesis/671051
Digital Biomarkers for Passive Remote Monitoring of Bipolar Disorder: Systematic Review (Preprint)Preprint
Kutcher, T. P., Chakraborty, I., Kostick, K. M., Sano, A., Moukaddam, N. J., Herron, J. A., Goodman, W. K., Sheth, S. A., Sabharwal, A., & Provenza, N. R. (2025) · doi:10.2196/preprints.84675
Whitmore, N. W., Shimizu, K., Shifat, M. N. U. H., & Maes, P. (2026) · doi:10.2196/preprints.98561
Short, N. A., An, X., Ji, Y., Li, Q., Neylan, T. C., Clifford, G. D., House, S. L., Beaudoin, F. L., Stevens, J. S., Linnstaedt, S. D., et al. (2026) · Journal of Medical Internet Research · doi:10.2196/73771
“Research in this area is in nascent stages but suggests that keystroke behavior data, alone [31-33] or in combination with other smartphone data [34], are useful for identifying increased depressive symptoms and cognitive functioning/impairments [35,36] among individuals with depression and bipolar…”
Leveraging federated learning for mental healthcare: a systematic review
Warrier, L. C., & K, R. G. (2026) · Multimedia Tools and Applications · doi:10.1007/s11042-026-21679-8 · Cited by 1
Digital Markers for Passive Remote Monitoring of Bipolar Disorder: Systematic Review
Kutcher, T. P., Chakraborty, I., Kostick, K. M., Sano, A., Moukaddam, N. J., Herron, J. A., Goodman, W. K., Sheth, S. A., Sabharwal, A., & Provenza, N. R. (2026) · JMIR Mental Health · doi:10.2196/94819 · Cited by 2
“Using the same BiAffect typing dataset, Cao et al [60] applied a deep learning architecture using gated recurrent units to classify bipolar depression with 90% accuracy.”
Parsing inter-individual variability in the digital phenotype across the menstrual cyclePreprint
Knol, L., Nagpal, A., Hussain, F., Beckmann, C. F., Leow, A., Eisenlohr-Moul, T. A., & Marquand, A. F. (2026) · medRxiv · doi:10.64898/2026.08.26.26361403
Self-citationLee, E., Kim, S., Im, S., Oh, H., & Kang, S. (2026) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3831648 · Cited by 1
Metadata
- Authors
- Cao, B., Zheng, L., Zhang, C., Yu, P. S., Piscitello, A., Zulueta, J., Ajilore, O., Ryan, K., & Leow, A. D.
- Affiliations
- University of Illinois Chicago (US)
- University of Illinoisat Chicago, Chicago, IL, USA
- University of Michigan (US)
- ORCID
- Venue
- Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
- Publisher
- ACM
- Published
- 2017-08-04
- Pages
- 747-755
- OpenAlex
- W2744939564
- Cited by
- 145
- References
- 61
- Citations by year
- 2026: 6 · 2025: 9 · 2024: 11 · 2023: 10 · 2022: 21 · 2021: 28 · 2020: 23 · 2019: 20 · 2018: 17
- FWCI
- 18.87
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Science Foundation (IIS-1526499 and CNS-1626432)
- National Natural Science Foundation of China (61672313)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Emotion and Mood Recognition
- Social Sciences › Psychology › Social Psychology › Mental Health via Writing
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Mental Health Research Topics
- Keywords
- mood prediction, mobile phone metadata, bipolar disorder, depression severity, decision fusion, multi-view fusion, mental health monitoring
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
