A Ubiquitous Clinic Recommendation System Using the Modified Mixed-Binary Nonlinear Programming–Feedforward Neural Network Approach

Most of the existing ubiquitous clinic recommendation (UCR) systems adopt linear mechanisms to aggregate the attribute-level performances of a clinic to evaluate the overall performance. However, such linear mechanisms may not be able to explain the choices of all patients. To solve this problem, the modified mixed binary nonlinear programming (MMBNLP)–feedforward neural network (FNN) approach is proposed in this study. In the proposed methodology, first, the existing MBNLP model is modified to improve the successful recommendation rate using a linear recommendation mechanism. Subsequently, an FNN is constructed to fit the relationship between the attribute-level performances of a clinic and its overall performance, thereby providing possible ways to further enhance the recommendation performance. The results of a regional experiment showed that the MMBNLP–FNN approach improved the successful recommendation rate by 30%.

Leveraging End-User Data for Enhanced Design Concept Evaluation: A Multimodal Deep Regression Model

Design concept evaluation is a key process in the new product development process with a significant impact on the product's success and total cost over its life cycle. This paper is motivated by two limitations of the state-of-the-art in concept evaluation: (1) The amount and diversity of user feedback and insights utilized by existing concept evaluation methods such as quality function deployment are limited. (2) Subjective concept evaluation methods require significant manual effort which in turn may limit the number of concepts considered for evaluation. A Deep Multimodal Design Evaluation (DMDE) model is proposed in this paper to bridge these gaps by providing designers with an accurate and scalable prediction of new concepts' overall and attribute-level desirability based on large-scale user reviews on existing designs. The attribute-level sentiment intensities of users are first extracted and aggregated from online reviews. A multimodal deep regression model is then developed to predict the overall and attribute-level sentiment values based on the features extracted from orthographic product images via a fine-tuned ResNet-50 model and from product descriptions via a fine-tuned BERT model, and aggregated using a novel self-attention-based fusion model. The DMDE model adds a data-driven, user-centered loop within the concept development process to better inform the concept evaluation process. Numerical experiments on a large dataset from an online footwear store indicate a promising performance by the DMDE model with 0.001 MSE loss and over 99.1% accuracy.

Disentangled Face Attribute Editing via Instance-Aware Latent Space Search

Recent works have shown that a rich set of semantic directions exist in the latent space of Generative Adversarial Networks (GANs), which enables various facial attribute editing applications. However, existing methods may suffer poor attribute variation disentanglement, leading to unwanted change of other attributes when altering the desired one. The semantic directions used by existing methods are at attribute level, which are difficult to model complex attribute correlations, especially in the presence of attribute distribution bias in GAN's training set. In this paper, we propose a novel framework (IALS) that performs Instance-Aware Latent-Space Search to find semantic directions for disentangled attribute editing. The instance information is injected by leveraging the supervision from a set of attribute classifiers evaluated on the input images. We further propose a Disentanglement-Transformation (DT) metric to quantify the attribute transformation and disentanglement efficacy and find the optimal control factor between attribute-level and instance-specific directions based on it. Experimental results on both GAN-generated and real-world images collectively show that our method outperforms state-of-the-art methods proposed recently by a wide margin. Code is available at https://github.com/yxuhan/IALS.

Attribute-wise Explainable Fashion Compatibility Modeling

With the boom of the fashion market and people’s daily needs for beauty, clothing matching has gained increased research attention. In a sense, tackling this problem lies in modeling the human notions of the compatibility between fashion items, i.e., Fashion Compatibility Modeling (FCM), which plays an important role in a wide bunch of commercial applications, including clothing recommendation and dressing assistant. Recent advances in multimedia processing have shown remarkable effectiveness in accurate compatibility evaluation. However, these studies work like a black box and cannot provide appropriate explanations, which are indeed of importance for gaining users’ trust and improving their experience. In fact, fashion experts usually explain the compatibility evaluation through the matching patterns between fashion attributes (e.g., a silk tank top cannot go with a knit dress). Inspired by this, we devise an attribute-wise explainable FCM solution, named ExFCM , which can simultaneously generate the item-level compatibility evaluation for input fashion items and the attribute-level explanations for the evaluation result. In particular, ExFCM consists of two key components: attribute-wise representation learning and attribute interaction modeling. The former works on learning the region-aware attribute representation for each item with the threshold global average pooling. Besides, the latter is responsible for compiling the attribute-level matching signals into the overall compatibility evaluation adaptively with the attentive interaction mechanism. Note that ExFCM is trained without any attribute-level compatibility annotations, which facilitates its practical applications. Extensive experiments on two real-world datasets validate that ExFCM can generate more accurate compatibility evaluations than the existing methods, together with reasonable explanations.

Conserving rhinos by legal trade: Insights from a choice experiment on rhino horn consumers

A legal rhino horn trade is suggested to reduce poaching. To examine this proposition we conducted a choice experiment with 345 rhino horn consumers in Vietnam investigating their preferences for legality, source, price and peer experience of medicinal efficacy as attributes in their decision to purchase rhino horn. We calculated consumers’ willingness to pay for each attribute level. Consumers preferred and were willing to pay more for wild than semi-wild and farmed rhino horn but showed the strongest preference for legal horn although higher-income consumers were less concerned about legality. The number of peers having used rhino horn without positive effect reduced preference for wild-sourced horn and increased preference for legality. Hence, a legal trade in rhino horn would likely not eliminate a parallel black market. Whether poaching would be reduced depends on the price difference in the two markets, campaigns ability to change consumer preferences, and regulation efforts.

Efficient sensitivity orient blockchain encryption for improved data security in cloud

Data security in the cloud has become a dominant topic being discussed in recent times as the security of data in the cloud has been focused on by several researchers. However, the data security was enforced at the attribute level, the adversaries are capable of learning the method of data encryption even there are access restrictions are enforced at an attribute level. To challenge the adversaries with more sophisticated security measures, an efficient real-time service-centric feature sensitivity analysis (RSFSA) model is proposed in this paper. The RSFSA model analyses the sensitivity of different features being accessed by any service and at multiple levels. At each level, the method checks the set of features being accessed and the number of features the user has access grant to compute the FLAG value for the user according to the profile given. Based on the value of FLAG, the user has been granted or denied service access. On the other side, the method maintains different encryption schemes and keys for each level of features. As the features are organized in multiple levels, the method maintains a set of schemes and keys for each level dedicative. Based on the service level and data, the method selects an encryption scheme and key to perform data encryption. According to that, the service access data has been encrypted at the attribute level with a specific scheme and key. Data encrypted has been uploaded to the blockchain and the method modifies the reference part of the chain to connect only the blocks to which the user has access. The chain given to the user would do not contain any reference from a specific block to which the user has no access. The proposed method improves the performance of data security and access restriction greatly.

AKC-Based Revocable ABE Schemes from LWE Assumption

The emergence of quantum computing threatens many classical cryptographic schemes, leading to the innovations in public-key cryptography for postquantum cryptography primitives and protocols that resist to quantum attacks. Lattice-based cryptography is considered to be one of the promising mathematical approaches to achieving security resistant to quantum attacks, which could be built on the learning with errors (LWE) problem and its variants. The fundamental building blocks of protocols for public-key encryption (PKE) and key encapsulation mechanism (KEM) submitted to the National Institute of Standards and Technology (NIST) based on LWE and its variants are called key consensus (KC) and asymmetric key consensus (AKC) by Jin et al. They are powerful tools for constructing PKE schemes. In this work, we further demonstrate the power of KC/AKC by proposing two special types of PKE schemes, namely, revocable attribute-based encryption (RABE). To be specific, on the basis of AKC and PKE/KEM protocols submitted to the NIST based on LWE and its variants, combined with full-rank difference, trapdoor on lattices, sampling algorithms, leftover hash lemma, and binary tree structure, we propose two directly revocable ciphertext-policy attribute-based encryption (DR-ABE) schemes from LWE, which support flexible threshold access policies on multivalued attributes, achieving user-level and attribute-level user revocation, respectively. Specifically, the construction of the ciphertext is derived from AKC, and the revocation list is defined and embedded into the ciphertext by the message sender to revoke a user in the user-level revocable scheme or revoke some attributes of a certain user in the attribute-level revocable scheme. We also discuss how to outsource decryption and reduce the workload for the end user. Our schemes proved to be secure in the standard model, assuming the hardness of the LWE problem. The two schemes imply the versatility of KC/AKC.