RETRACTED ARTICLE: Context-Category Specific sequence aware Point-Of-Interest Recommender System with Multi-Gated Recurrent Unit

Kala, K. U.; Nandhini, M.

doi:10.1007/s12652-019-01583-w

Cited by 10 publications

(21 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Foursquare: Acc@10=0.8851, Yelp: Acc@10=0.5587, Brightkite: Acc@10=0.7385 MGRU, JAIHC 2019 [47] Capture dynamic and transition context using Multi-GRU (Two special gate are added with GRU).…”

Section: Methods Summery Performancementioning

confidence: 99%

“…For capturing different contextual impact on the users' preferences this model uses two types of gating mechanisms i.e., (1) Contextual Attention Gate (CAG): controls the influence of ordinary and transition contexts on the users' dynamic preferences and (2) Time-and Spatial-based Gate (TSG): considers the time intervals and geographical distances between successive check-ins to control the influence of the hidden state of previous GRU units. Kala et al proposed [47] Multi-GRU (MGRU) which modifies the basic GRU unit by adding two additional gates for a better recommendation. The first added gate is Dynamic Contextual-Attention-Gate (DCAG-α) which captures the effect of dynamic contexts like -time of the day, companion, user's mood, etc.…”

Section: Grumentioning

confidence: 99%

“…Basic RNN -2016 ST-RNN [37] 2018 MCI-DNN [73] 2020 Flashback [38], ASPPA [39], LLRec [40], DRLM [72] LSTM Basic LSTM 2018 HST-LSTM [74], TMCA [42], SLSTM [83] 2019 LSPL [75], ASTEN [77], ATST-LSTM [80], MMR [81], SGBA [82] 2020 PLSPL [76], LSTPM [41], iMTL [43], CatDM [44], ARNN [78], STAR [79] Bi-LSTM 2020 GT-HAN [84], t-LocPred [86], CAPRE [87] Modified LSTM 2020 STGN [88] Self-Attention 2020 GeoSAN [48], SANST [49] GRU -2018 DeepMove [45], CARA [46] 2019 MGRU [47] Graph Embedding -2016 GE [90] 2017 STA [50] 2018 JLGE [92] 2019 RELINE [51] 2020 DYSTAL [52], HCT [96], UP2VEC [98], HMRM [97] GAN -2019 Geo-ALM [99], APOIR [100] 2020 AdattTUL [101] Others -2017 Geo-Teaser [23], LCE [114] 2018 CAPE [102], ST-DME [115] 2019 Bi-STDDP [103], TEMN…”

Section: Category Subcategory Year Referencementioning

confidence: 99%

“…Data sparsity is one of the most crucial problems of building a location recommender system because the graphs and matrices are far more sparse than most other recommender systems. From a user's point of view, a person travels to very few locations in his/her lifetime compared to the sheer huge number of [88], LSTPM [41], TMCA [42], ASTEN [77], Bi-STDDP [103], GT-HAN [85], Flashback [38], TEMN [104], ST-RNN [37], HST-LSTM [74], DeepMove [45], SANST [49] ✓ ✓ ✗ ✗ ✓ LSPL [75], PLSPL [76], iMTL [43], ASPPA [39], ARNN [78], CARA [46], GT-HAN [84], t-LocPred [86], CatDM [44], GE [90], MGRU [47] [48], Geo-Teaser [23] ✓ ✓ ✗ ✗ ✗ CAPRE [87] ✓ ✓ ✓ ✓ ✗ STA [50] ✗ ✓ ✓ ✗ ✓ JLGE [92], RELINE [51], APOIR [100] ✓ ✓ ✗ ✓ ✓ LLRec (Teacher) [40] ✓ ✓ ✓ ✓ ✓ HME [109] ✓ ✓ ✓ ✗ ✗ DYSTAL [52] ✗ ✓ ✓ ✓ ✓ HCT [96] ✗ ✓ ✓ ✗ ✗…”

Section: Data Sparsitymentioning

confidence: 99%

“…In the last few years, we have seen an unprecedented rise in the number of works leveraging deep learning in POI recommendation in all major venues (e.g., AAAI, IJCAI, SIGIR, CIKM, WWW, etc.). The use of different deep learning paradigms such as CNN [35,36], RNN [37,38,39,40], Long Short Term Memory (LSTM) [41,42,43,44], Gated Recurrent Unit (GRU) [45,46,47], and self-attention [48,49] have greatly boosted the performance of POI recommendation models. On top of that, state-of-the-art techniques from Natural Language Processing (NLP) have also been employed for complex modeling of human mobility in POI recommendation.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Survey on Deep Learning Based Point-Of-Interest (POI) Recommendations

Islam¹,

Mohammad²,

Das³

et al. 2020

Preprint

View full text Add to dashboard Cite

Location-based Social Networks (LBSNs) enable users to socialize with friends and acquaintances by sharing their check-ins, opinions, photos, and reviews. Huge volume of data generated from LBSNs opens up a new avenue of research that gives birth to a new sub-field of recommendation systems, known as Point-of-Interest (POI) recommendation. A POI recommendation technique essentially exploits users' historical check-ins and other multi-modal information such as POI attributes and friendship network, to recommend the next set of POIs suitable for a user. A plethora of earlier works focused on traditional machine learning techniques by using hand-crafted features from the dataset. With the recent surge of deep learning research, we have witnessed a large variety of POI recommendation works utilizing different deep learning paradigms. These techniques largely vary in problem formulations, proposed techniques, used datasets and features, etc. To the best of our knowledge, this work is the first comprehensive survey of all major deep learning-based POI recommendation works. Our work categorizes and critically analyzes the recent POI recommendation works based on different deep learning paradigms and other relevant features. This review can be considered a cookbook for researchers or practitioners working in the area of POI recommendation.

show abstract

Section: Methods Summery Performancementioning

confidence: 99%

Section: Grumentioning

confidence: 99%