{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T09:27:31Z","timestamp":1775813251106,"version":"3.50.1"},"reference-count":73,"publisher":"Institute for Operations Research and the Management Sciences (INFORMS)","issue":"4","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Management Science"],"published-print":{"date-parts":[[2026,4]]},"abstract":"<jats:p>The recent surge in the popularity of short-form video (SFV) on digital platforms has led to massive numbers of videos and ever-evolving topics. As a result, the task of making personalized recommendations has become increasingly challenging. We introduce a new pure exploration problem on SFV platforms: finding a ([Formula: see text])-optimal set that includes all recommendations within the [Formula: see text]-optimality gap and that excludes those beyond the [Formula: see text]-optimality gap relative to the best arm with a capacity limit of K. To solve this problem, we propose an algorithm called adaptive acquisition tree (AAT). AAT jointly accounts for user preference heterogeneity and high-dimensional product characteristics. It adaptively segments users and then, learns a personalized transductive policy that can be used on partially observed or even unobserved card types to accommodate the dynamic trends on SFV platforms. We derive the sample complexity required to identify a [Formula: see text]-optimal set. Our method\u2019s efficiency is validated through numerical tests using data from the NetEase platform. Our results reveal that the proposed policy performs significantly better than several state-of-the-art benchmarks across four transductive scenarios for both spotlight recommendations (i.e., best-arm identifications) and [Formula: see text]-optimal set recommendations. Compared with the best benchmarks for the best card and [Formula: see text]-optimal set recommendations, our approach can elevate the average rewards (measured by view time) by 30% (to 100%) and 43% (to 56%), respectively. Given the increasing popularity and uniqueness of SFVs and more broadly, user-generated content, our method offers significant academic and practical merit.<\/jats:p>\n                  <jats:p>This paper was accepted by Omar Besbes, revenue management and market analytics.<\/jats:p>\n                  <jats:p>Funding: Y. Leng is supported by the U.S. National Science Foundation (NSF) under [Grant IIS 2153468].<\/jats:p>\n                  <jats:p>Supplemental Material: The online appendix and data files are available at https:\/\/doi.org\/10.1287\/mnsc.2022.01130 .<\/jats:p>","DOI":"10.1287\/mnsc.2022.01130","type":"journal-article","created":{"date-parts":[[2025,8,25]],"date-time":"2025-08-25T15:53:32Z","timestamp":1756137212000},"page":"3592-3610","source":"Crossref","is-referenced-by-count":1,"title":["Adaptive Data Acquisition for Personalized Recommendations with Optimality Guarantees on Short-Form Video Platforms"],"prefix":"10.1287","volume":"72","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9235-1411","authenticated-orcid":false,"given":"Junyu","family":"Cao","sequence":"first","affiliation":[{"name":"McCombs School of Business, The University of Texas at Austin, Austin, Texas 78712"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7084-2700","authenticated-orcid":false,"given":"Yan","family":"Leng","sequence":"additional","affiliation":[{"name":"McCombs School of Business, The University of Texas at Austin, Austin, Texas 78712"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"109","reference":[{"key":"B1","volume":"29","author":"Abernethy JD","year":"2016","journal-title":"Adv. Neural Inform. Processing Systems"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1287\/msom.2023.1195"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1007\/s00186-021-00744-y"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1111\/poms.12963"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1287\/msom.2018.0725"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-75729-8_7"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1287\/msom.2021.1047"},{"issue":"5","key":"B8","first-page":"2095","volume":"65","author":"Bernstein F","year":"2019","journal-title":"Management Sci."},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1287\/msom.2015.0548"},{"key":"B10","unstructured":"Bubeck S, Wang T, Viswanathan N (2013) Multiple identifications in multi-armed bandits.\n                      Internat. Conf. Machine Learn\n                      . (PMLR, New York), 258\u2013265."},{"key":"B11","doi-asserted-by":"crossref","unstructured":"Cao J, Sun W (2019) Dynamic learning of sequential choice bandit problem under marketing fatigue.\n                      Proc. AAAI Conf. Artificial Intelligence\n                      33(1):3264\u20133271.","DOI":"10.1609\/aaai.v33i01.33013264"},{"key":"B12","doi-asserted-by":"crossref","unstructured":"Cao J, Sun W, Shen Z-JM (2019) Doubly adaptive cascading bandits with user abandonment. Preprint, submitted April 8, http:\/\/dx.doi.org\/10.2139\/ssrn.3355211.","DOI":"10.2139\/ssrn.3355211"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1198\/106186004X13064"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1287\/opre.2020.2016"},{"key":"B15","unstructured":"Chen L, Li J (2015) On the optimal sample complexity for best arm identification. Preprint, submitted November 12, https:\/\/arxiv.org\/abs\/1511.03774."},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1287\/mnsc.2021.4256"},{"key":"B17","doi-asserted-by":"crossref","unstructured":"Chen N, Gallego G, Tang Z (2019) The use of binary choice forests to model and estimate discrete choices. Preprint, submitted August 3, https:\/\/arxiv.org\/abs\/1908.01109.","DOI":"10.2139\/ssrn.3430886"},{"key":"B18","unstructured":"Chen L, Li J, Qiao M (2017b) Nearly instance optimal sample complexity bounds for top-k arm selection.\n                      Proc. 20th Internat. Conf. Artificial Intelligence Statist.\n                      (PMLR, New York), 101\u2013110."},{"key":"B19","unstructured":"Chen J, Chen X, Zhang Q, Zhou Y (2017a) Adaptive multiple-arm identification.\n                      Internat. Conf. Machine Learn\n                      . (PMLR, New York), 722\u2013730."},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.1287\/mnsc.2022.4651"},{"key":"B21","volume":"27","author":"Chen S","year":"2014","journal-title":"Adv. Neural Inform. Processing Systems"},{"key":"B22","unstructured":"Cheung M-C (2020) Why short-form video apps are so popular in China.\n                      eMarketer\n                      (January 2), https:\/\/www.emarketer.com\/content\/why-short-form-video-apps-are-so-popular-in-china."},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1287\/opre.2017.1629"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1287\/opre.2022.2301"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1287\/mnsc.2021.4044"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.1214\/14-STS500"},{"key":"B27","unstructured":"Elmachtoub AN, McNellis R, Oh S, Petrik M (2017) A practical method for solving contextual bandit problems using decision trees. Preprint, submitted June 14, https:\/\/arxiv.org\/abs\/1706.04687."},{"issue":"39","key":"B28","first-page":"1079","volume":"7","author":"Even-Dar E","year":"2006","journal-title":"J. Machine Learn. Res."},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.1287\/msom.2015.0561"},{"key":"B30","volume":"33","author":"Fiez T","year":"2019","journal-title":"Adv. Neural Inform. Processing Systems"},{"key":"B31","volume":"25","author":"Gabillon V","year":"2012","journal-title":"Adv. Neural Inform. Processing Systems"},{"key":"B32","doi-asserted-by":"crossref","unstructured":"Gelada C, Bellemare MG (2019) Off-policy deep reinforcement learning by bootstrapping the covariate shift.\n                      Proc. AAAI Conf. Artificial Intelligence\n                      33(1):3647\u20133655.","DOI":"10.1609\/aaai.v33i01.33013647"},{"key":"B33","unstructured":"Gentile C, Li S, Zappella G (2014) Online clustering of bandits.\n                      Internat. Conf. Machine Learn\n                      . (PMLR, New York), 757\u2013765."},{"key":"B34","unstructured":"Gentile C, Li S, Kar P, Karatzoglou A, Zappella G, Etrue E (2017) On context-dependent clustering of bandits.\n                      Internat. Conf. Machine Learn.\n                      (PMLR, New York), 1253\u20131262."},{"key":"B35","doi-asserted-by":"crossref","unstructured":"Gretton A, Smola A, Huang J, Schmittfull M, Borgwardt K, Sch\u00f6lkopf B (2009) Covariate shift by kernel mean matching.\n                      Dataset Shift Machine Learn.\n                      (MIT Press, Cambridge, MA), 131\u2013160.","DOI":"10.7551\/mitpress\/9780262170055.003.0008"},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.1287\/msom.2017.0691"},{"key":"B37","doi-asserted-by":"publisher","DOI":"10.1287\/mnsc.1060.0619"},{"key":"B38","unstructured":"HubSpot (2025) The top video marketing tactics brands are investing in [+which are losing steam]. Accessed August 10, 2025, https:\/\/blog.hubspot.com\/marketing\/top-video-marketing-tactics?"},{"key":"B39","doi-asserted-by":"publisher","DOI":"10.1287\/opre.2018.1739"},{"key":"B40","first-page":"10007","volume":"33","author":"Jedra Y","year":"2020","journal-title":"Adv. Neural Inform. Processing Systems"},{"key":"B41","unstructured":"Jiang H, Li J, Qiao M (2017) Practical algorithms for best-\n                      K\n                      identification in multi-armed bandits. Preprint, submitted May 19, https:\/\/arxiv.org\/abs\/1705.06894."},{"key":"B42","unstructured":"Kalyanakrishnan S, Stone P (2010) Efficient selection of multiple bandit arms: Theory and practice.\n                      Proc. 27th Internat. Conf. Internat. Conf. Machine Learn.\n                      (Omnipress, Madison, WI)."},{"key":"B43","unstructured":"Karnin Z, Koren T, Somekh O (2013) Almost optimal exploration in multi-armed bandits.\n                      Internat. Conf. Machine Learn\n                      . (PMLR, New York), 1238\u20131246."},{"issue":"1","key":"B44","first-page":"1","volume":"17","author":"Kaufmann E","year":"2016","journal-title":"J. Machine Learn. Res."},{"key":"B45","doi-asserted-by":"publisher","DOI":"10.1016\/j.orl.2021.03.011"},{"key":"B46","doi-asserted-by":"publisher","DOI":"10.1287\/opre.2022.0112"},{"key":"B47","doi-asserted-by":"publisher","DOI":"10.1287\/isre.2018.0833"},{"key":"B48","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-005-0466-3"},{"key":"B49","doi-asserted-by":"publisher","DOI":"10.1017\/9781108571401"},{"key":"B50","doi-asserted-by":"publisher","DOI":"10.1287\/isre.2021.0343"},{"key":"B51","doi-asserted-by":"crossref","unstructured":"Leng Y, Ruiz R, Liu X (2020) Interpretable recommendations and user-centric explanations with geometric deep learning. Preprint, submitted November 13, https:\/\/dx.doi.org\/10.2139\/ssrn.3696092.","DOI":"10.2139\/ssrn.3696092"},{"key":"B52","doi-asserted-by":"publisher","DOI":"10.1287\/mnsc.2020.3827"},{"key":"B53","unstructured":"Lorenz T (2020) This is why you heard about TikTok so much in 2020.\n                      New York Times\n                      (December 31), https:\/\/www.nytimes.com\/2020\/12\/31\/style\/tiktok-trends-2020.html."},{"key":"B54","volume":"33","author":"Mason B","year":"2020","journal-title":"Adv. Neural Inform. Processing Systems"},{"key":"B55","doi-asserted-by":"publisher","DOI":"10.1177\/1527476420982230"},{"key":"B56","doi-asserted-by":"publisher","DOI":"10.1111\/poms.13783"},{"key":"B57","doi-asserted-by":"publisher","DOI":"10.1287\/opre.2019.1928"},{"key":"B58","doi-asserted-by":"crossref","unstructured":"Nguyen TT, Lauw HW (2014) Dynamic clustering of contextual multi-armed bandits.\n                      Proc. 23rd ACM Internat. Conf. Inform. Knowledge Management\n                      (ACM, New York), 1959\u20131962.","DOI":"10.1145\/2661829.2662063"},{"key":"B59","unstructured":"Quinlan JR (1992) Learning with continuous classes.\n                      Proc. 5th Australian Joint Conf. Artificial Intelligence\n                      , vol. 92 (World Scientific, Singapore), 343\u2013348."},{"key":"B60","unstructured":"Ren W, Liu J, Shroff NB (2019) Exploring\n                      k\n                      out of top \u03c1 fraction of arms in stochastic bandits.\n                      22nd Internat. Conf. Artificial Intelligence Statist.\n                      (PMLR, New York), 2820\u20132828."},{"key":"B61","doi-asserted-by":"publisher","DOI":"10.2307\/2600984"},{"key":"B62","unstructured":"Russo D (2016) Simple Bayesian algorithms for best arm identification.\n                      Conf. Learn. Theory\n                      (PMLR, New York), 1417\u20131418."},{"key":"B63","doi-asserted-by":"publisher","DOI":"10.1287\/mksc.2016.1023"},{"key":"B64","doi-asserted-by":"publisher","DOI":"10.1137\/1.9781611973433"},{"key":"B65","doi-asserted-by":"publisher","DOI":"10.1287\/isre.2019.0893"},{"key":"B66","volume":"27","author":"Soare M","year":"2014","journal-title":"Adv. Neural Inform. Processing Systems"},{"key":"B67","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/7503.003.0172"},{"key":"B68","unstructured":"Tao C, Blanco S, Zhou Y (2018) Best arm identification in linear bandits with linear dimension dependency.\n                      Internat. Conf. Machine Learn\n                      . (PMLR, New York), 4877\u20134886."},{"key":"B69","unstructured":"Wallaroo (2021) TikTok statistics. (May 7), https:\/\/wallaroomedia.com\/blog\/social-media\/tiktok-statistics\/."},{"key":"B70","unstructured":"Xu L, Honda J, Sugiyama M (2018) A fully adaptive algorithm for pure exploration in linear bandits.\n                      Internat. Conf. Artificial Intelligence Statist\n                      . (PMLR, New York), 843\u2013851."},{"key":"B71","doi-asserted-by":"crossref","unstructured":"Yang J, Liu C, Teng M, Liao M, Xiong H (2016) Buyer targeting optimization: A unified customer segmentation perspective.\n                      2016 IEEE Internat. Conf. Big Data\n                      (IEEE, Piscataway, NJ), 1262\u20131271.","DOI":"10.1109\/BigData.2016.7840730"},{"key":"B72","doi-asserted-by":"crossref","unstructured":"Yu K, Bi J, Tresp V (2006) Active learning via transductive experimental design.\n                      Proc. 23rd Internat. Conf. Machine Learn.\n                      (PMLR, New York), 1081\u20131088.","DOI":"10.1145\/1143844.1143980"},{"key":"B73","doi-asserted-by":"publisher","DOI":"10.1287\/msom.2020.0923"}],"container-title":["Management Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubsonline.informs.org\/doi\/pdf\/10.1287\/mnsc.2022.01130","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T08:28:55Z","timestamp":1775809735000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubsonline.informs.org\/doi\/10.1287\/mnsc.2022.01130"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4]]},"references-count":73,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2026,4]]}},"alternative-id":["10.1287\/mnsc.2022.01130"],"URL":"https:\/\/doi.org\/10.1287\/mnsc.2022.01130","relation":{},"ISSN":["0025-1909","1526-5501"],"issn-type":[{"value":"0025-1909","type":"print"},{"value":"1526-5501","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4]]}}}