An official website of the United States government
Recommending suitable tags for online textual content is a key building block for better content organization and consumption. In this paper, we identify three pillars that impact the accuracy of tag recommendation: (1) sequential text modeling meaning that the intrinsic sequential ordering as well as different areas of text might have an important implication on the corresponding tag(s) , (2) tag correlation meaning that the tags for a certain piece of textual content are often semantically correlated with each other, and (3) content-tag overlapping meaning that the vocabularies of content and tags are overlapped. However, none of the existing methods consider all these three aspects, leading to a suboptimal tag recommendation. In this paper, we propose an integral model to encode all the three aspects in a coherent encoder-decoder framework. In particular, (1) the encoder models the semantics of the textual content via Recurrent Neural Networks with the attention mechanism, (2) the decoder tackles the tag correlation with a prediction path, and (3) a shared embedding layer and an indicator function across encoder-decoder address the content-tag overlapping. Experimental results on three realworld datasets demonstrate that the proposed method significantly outperforms the existing methods in terms of recommendation accuracy.
more »
« less
Guiding supervised topic modeling for content based tag recommendation
Automatically recommending suitable tags for online content is a necessary task for better information organization and retrieval. In this article, we propose a generative model SimWord for the tag recommendation problem on textual content. The key observation of our model is that the tags and their relevant/similar words may have appeared in the corresponding content. In particular, we first empirically verify this observation in real data sets, and then design a supervised topic model which is guided by the above observation for tag recommendation. Experimental evaluations demonstrate that the proposed method outperforms several existing methods in terms of recommendation accuracy.
more »
« less
- PAR ID:
- 10099231
- Date Published:
- Journal Name:
- Neurocomputing
- ISSN:
- 0925-2312
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Passive radio-frequency identification (RFID) tags are attractive because they are low cost, battery-free, and easy to deploy. This technology is traditionally being used to identify tags attached to the objects. In this paper, we explore the feasibility of turning passive RFID tags into battery-free temperature sensors. The impedance of the RFID tag changes with the temperature and this change will be manifested in the reflected signal from the tag. This opens up an opportunity to realize battery-free temperature sensing using a passive RFID tag with already deployed Commercial Off-the-Shelf (COTS) RFID reader-antenna infrastructure in supply chain management or inventory tracking. However, it is challenging to achieve high accuracy and robustness against the changes in the environment. To address these challenges, we first develop a detailed analytical model to capture the impact of temperature change on the tag impedance and the resulting phase of the reflected signal. We then build a system that uses a pair of tags, which respond differently to the temperature change to cancel out other environmental impacts. Using extensive evaluation, we show our model is accurate and our system can estimate the temperature within a 2.9 degree centigrade median error and support a normal read range of 3.5 m in an environment-independent manner.more » « less
-
Abstract Achieving long‐term retention of pop‐up satellite archival tags (PSATs) has proven difficult for all fishes but is particularly challenging for small migrant species due to the relatively large size of tags. In this study, the authors tested the latest and smallest PSAT model on the market, the mark‐report satellite tag (mrPAT), and developed a simple, cost‐effective method of tag attachment on sheepshead Archosargus probatocephalus (Walbaum 1792), a small marine fish. During laboratory trials, the method of tag attachment used in this study outperformed the existing methods with two c . 40 cm fish retaining their tags for 3 months (the duration of the laboratory study). During field deployments, data were successfully obtained for 17 of the 25 tagged fish [37–50 cm fork length (FL)]. Of these, 14 tags (82%) remained on the fish until the pre‐programmed release date resulting in tag retention times of up to 172 days (mean: 140 days). The investigation represents the first extensive study into the feasibility of PSATs for monitoring fishes in this size range. The authors demonstrate that their method of attachment and this latest PSAT model are feasible for c . 5‐month deployments on fishes that are relatively small ( c . 45 cm FL). These results with A. probatocephalus represent a potentially significant advance in PSAT methodology for fishes of this size. Future investigations are needed to determine if this method is transferrable to other species in the same size range.more » « less
-
Radio frequency identification (RFID) is a technology for automated identification of objects and people. RFID technology is expected to find extensive use in applications related to the Internet of Things, and in particular applications of Internet of Battlefield Things. Of particular interest are passive RFID tags due to a number of their salient advantages. Such tags, lacking energy sources of their own, use backscattering of the power of an RF source (a reader) to communicate. Recently, passive RFID tag-to-tag (T2T) communication has been demonstrated, via which tags can directly communicate with each other and share information. This opens the possibility of building a Network of Tags (NeTa), in which the passive tags communicate among themselves to perform data processing functions. Among possible applications of NeTa are monitoring services in hard-to-reach locations. As an essential step toward implementation of NeTa, we consider a novel multi-hop network architecture; in particular, with the proposed novel turbo backscattering operation, inter-tag distances can be significantly increased. Due to the interference among tags’ transmissions, one of the main technical challenges of implementing such the NeTa architecture is the routing protocol design. In this paper, we introduce a design of a routing protocol, which is based on a solution of a non-linear binary optimization problem. We study the performance of the proposed protocol and investigate impacts of several network factors, such as the tag density and the transmit power of the reader.more » « less
-
Using tags within a mark-recapture framework allows researchers to assess population size and connectivity. Such methods have been applied in coastal zone habitats to monitor salt marsh restoration success by comparing the movement patterns of Mummichogs (Fundulus heteroclitus) between restored and natural marshes. Visible Implant Elastomer (VIE) tags are commonly used to tag small fish like Mummichogs, though the retention and survival of small fish using this method varies between studies, producing uncertainty during mark-recapture-based approaches. To address this, we conducted a laboratory experiment to determine the rate of tag loss and mortality of VIE tags on Mummichogs of two size classes (greater or less than 61 mm) and across different taggers. Tag loss and mortality increased over time, and the latter significantly varied between taggers. We then developed a predictive model, R package ‘retmort’, to account for the effect of this increase on mark-recapture studies. When adapted to a series of published works, our model provided rational estimates of tagging error for multiple species and tagging methods. Of the case studies the model was applied to (n = 26), 15 resulted in a percent standard error greater than 5%, signaling a significant percent of error due to uncounted, tagged animals. By not accounting for these individuals, recapture studies, particularly those that assess restoration efforts and coastal resilience, could underestimate the effects of those projects, leading to superfluous restoration efforts and erroneous recapture data for species with low tag retention and high mortality rates.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1016/j.neucom.2018.07.011
