Friend Recommendation Research Articles

Friend Recommendation using Unsupervised Machine Learning

Friend recommendation is one of a lot of accepted characteristics of amusing arrangement platforms, which recommends agnate or accustomed humans to users. The abstraction of friend recommendation originates from amusing networks such as Twitter and Facebook, which uses friends-of-friends adjustment to acclaim people. We can say users do not accomplish accompany from accidental humans but end up authoritative accompany with their friends’ friends. The absolute methods accept attenuated ambit of recommendation and are beneath efficient. Here in our proposed access, we are applying an added hierarchical clustering technique with the collaborative clarification advocacy algorithm as well the Principle Component Analysis (PCA) adjustment is activated for abbreviation the ambit of abstracts to get added accurateness in the results. The hierarchical clustering will accommodate added allowances of the clustering technique over the dataset, and the PCA will adviseredefining the dataset by abbreviating the ambit of the dataset as required. By implementing the above appearance of these two techniques on the acceptable collaborative clarification advocacy algorithm, the above apparatus acclimated for recommendations can be improved.

FAKTOR YANG MEMPENGARUHI KONSUMEN UNTUK MEMILIH 212 MART SURABAYA SEBAGAI TEMPAT UNTUK BERBELANJA

This research aims to find out factors that influence consumers decision on shopping at sharia mart 212 Mart at Wonorejo, Surabaya in the middle of the rapid growth of 212 Mart’s franchise around Indonesia. The method used in this research is explorative factor analysis. The population’s charateristics used in this study are consumers of 212 Mart Surabaya and people that live around Wonorejo that ever trade at 212 Mart. The amount of sample that are used in this study are 30 respondents for the pre-eliminary test and 95 respondents for the main test. This research used convenience sampling to withdraw samples. This research found that there are five factors that influence consumers decision on shopping at sharia mart 212 Mart Surabaya. The factors are consumer compability of halal goods factor, recommendation from friends that has the same vision about 212 movement factor, inventory factor, parking area factor and Islamic atsmosphere factor.Keywords: Consumers Decisions, Sharia Mart, Compability of halal goods, Friend Recommendation, inventory, Parking Area, Islamic Atsmosphere.

SEASONALITY OF LITHUANIAN NATURE TOURISM: VISITOR‘S PREFERENCES

Seasonality in the tourism sector is described as a change in tourism due to seasonal changes or a temporary change in tourist flows. In Lithuania, the tourism seasonality is recognized as a problem requiring a solution. This study has identified that every visitor/traveller who likes nature, prefers nature-based tourism or has plans to spend some time in nature is facing the tourism seasonality in Lithuania. The results of this survey support the idea that the most popular time for nature trips in Lithuania is summer time or the warm light season of the year. There are several reasons why visitors wish to go out to nature: many visitors hope to get good quality and exciting entertainment or educations, others are expecting for nature's aesthetic and other values, including beautiful landscapes and fresh air. Lithuanians usually choose nature trips with family or friends travelling by car for an average 3-7 days. The most of survey respondents plan trips in advance, search for information on the Internet, social networks, or use the recommendations of friends to visit one or another natural object, and almost do not use the services of tourism companies. Interest in one or another natural object and its attendance is primarily determined by weather conditions, activities and educations adapted to the exact season. This survey has shown that people wishing to choose non-seasonal natural tourism facing many problems. Depending on the season, tourism infrastructure could be more suited to all seasons, as well as respondents expect new educations adapted to different seasons or their indoor alternatives. In the recent years, more frequent extreme climate periods (heat, rainfall in summer, cold or wet periods in winter) have begun to change, most often reduce, visitor flows.

THE EFFECTIVENESS OF SOURCES OF INFORMATION ON RURAL FACILITIES AND AGRITOURISM

The aim of this dissertation was to diagnose the tendency of changes observed in the forms of information and promotion used by agricultural service providers and the sources of information preferred by tourists who rested in rural areas in 2001-2019. Moreover, the point of this study is to evaluate the effectiveness of sources of information and promotion on rural facilities used by service providers in agritourism. The empiric studies handled by the author of the article as well as by other researches, data from different institutions and literature formed the basis for this article. A comparative analysis was used to perform the research. It should be noticed that, in every period of the study, the forms of promotion used by agricultural service providers failed to fully meet tourists’ needs and expectations. In 2003, service providers believed in printed materials such as leaflets, pamphlets and catalogues as a form of advertisement and promotion, whereas tourists chose their destination to stay and relax at a rural facility relying on word-of-mouth marketing from friends, family and acquaintances. In 2012 and 2019 both parties pointed out the website as a prime source of information and the recommendations of friends and family in second place. In every empiric period, exchanges, fairs, advertising in catalogues and pamphlets as well as information in press releases, adverts on the radio or commercials in television were more important for service providers than for tourists. In order to meet tourists’ expectations, farmers and village residents should take care of the standard and design of their websites. To gain knowledge and necessary help on that matter, service providers may get in touch with agricultural advisory services or representatives of agritourism associations.

Dual Implicit Mining-Based Latent Friend Recommendation

The latent friend recommendation in online social media is interesting, yet challenging, because the user-item ratings and the user–user relationships are both sparse. In this paper, we propose a new dual implicit mining-based latent friend recommendation model that simultaneously considers the implicit interest topics of users and the implicit link relationships between the users in the local topic cliques. Specifically, we first propose an algorithm called all reviews from a user and all tags from their corresponding items to learn the implicit interest topics of the users and their corresponding topic weights, then compute the user interest topic similarity using a symmetric Jensen–Shannon divergence. After that, we adopt the proposed weighted local random walk with restart algorithm to analyze the implicit link relationships between the users in the local topic cliques and calculate the weighted link relationship similarity between the users. Combining the user interest topic similarity with the weighted link relationship similarity in a unified way, we get the final latent friend recommendation list. The experiments on real-world datasets demonstrate that the proposed method outperforms the state-of-the-art latent friend recommendation methods under four different types of evaluation metrics.

Social Link Inference via Multiview Matching Network From Spatiotemporal Trajectories.

In this article, we investigate the problem of social link inference in a target location-aware social network (LSN), which aims at predicting the unobserved links between users within the network. This problem is critical for downstream applications, including network completion and friend recommendation. In addition to the network structures commonly used in general link prediction, the studies tailored for social link inference in an LSN leverage user trajectories from the spatial aspect. However, the temporal factor lying in user trajectories is largely overlooked by most of the prior studies, limiting the capabilities of capturing the temporal relevance between users. Moreover, effective user matching by fusing different views, i.e., social, spatial, and temporal factors, remains unresolved, which hinders the potential improvement of link inference. To this end, this article devises a novel multiview matching network (MVMN) by regarding each of the three factors as one view of any target user pair. MVMN enjoys the flexibility and completeness of modeling each factor by developing its suitable matching module: 1) location matching module; 2) time-series matching module; and 3) relation matching module. Each module learns a view-specific representation for matching, and MVMN fuses them for final link inference. Extensive experiments on two real-world data sets demonstrate the superiority of our approach against several competitive baselines for link prediction and sequence matching, validating the contribution of its key components.

ATFR: Attribute based Trustworthy Friend Recommendations Online

Users today relay on social networking platforms for infor-mation retrieval. Information shared on the social networks will be cir-culated through their friend sharing them on their pro les. Thus making trustworthy friends in social networks is more important to maintain data privacy. Existing system compute friends based on their mutual friends which would not be e cient way of selecting friends online [1]. This allows user to get connected with strangers without any information of real friend or fake friend. To overcome this problem attribute based trustworthy friend recommendation system is proposed. Based on this, the attributes of each user is collected during pro le creation and these attributes are subjected to similarity between the users. Each user online are evaluated with their trust score with the help of Trust Score algo-rithm. Experimental result shows the number of friends recommended are trustworthy.

A semantic and social‐based collaborative recommendation of friends in social networks

SummaryThe development of social media technologies has greatly enhanced social interactions. The proliferation of social platforms has generated massive amounts of data and a considerable number of persons join these platforms every day. Therefore, one of the current issues is to facilitate the search for the most appropriate friends for a given user. We focus in this article on the recommendation of users in social networks. We propose a novel approach which combines a user‐based collaborative filtering (CF) algorithm with semantic and social recommendations. The semantic dimension suggests the close friends based on the calculation of the similarity between the active user and his friends. The social dimension is based on some social‐behavior metrics such as friendship and credibility degree. The novelty of our approach concerns the modeling of the credibility of the user, through his/her trust and commitment in the social network. A social recommender system based on this approach is developed and experiments have been conducted using the Yelp social network. The evaluation results demonstrated that the proposed hybrid approach improves the accuracy of the recommendation compared with the user‐based CF algorithm and solves the sparsity and cold start problems.

Vehicular Ad Hoc Network Representation Learning for Recommendations in Internet of Things

With the advancement of Internet of Things technology, we are able to collect massive people's trajectory data from various GPS services. These large amounts of trajectory records enable us to better understand human mobility patterns. Meanwhile, we are able to extract social relationships based on these digital records to provide personalized recommendation services, such as points of interests (POI) recommendation and friend recommendation. In this paper, we propose to recommend friends for taxi drivers based on vehicular trajectory records. For this purpose, we propose to construct a vehicular ad hoc network based on co-occurrence phenomenon. Furthermore, we take advantages of the network representation learning technique on the vehicular ad hoc network for learning driver vectors. Finally, potential friends are recommended based on the similarity of driver vectors. Extensive experimental results on two real-world datasets demonstrate that our proposed method has the best performance on friend recommendation compared with several state-of-the-art methods. To the best of our knowledge, this is the first attempt to recommend friends for taxi drivers based on vehicular ad hoc network representation learning.

Will Online Digital Footprints Reveal Your Relationship Status?

With the increasing social acceptance and openness, more and more sexual-minority men (SMM) have succeeded in creating and sustaining steady relationships in recent years. Maintaining steady relationships is beneficial to the wellbeing of SMM both mentally and physically. However, the relationship maintaining for them is also challenging due to the much less supports compared to the heterosexual couples, so that it is important to identify those SMM in steady relationship and provide corresponding personalized assistance. Furthermore, knowing SMM's relationship and the correlations with other visible features is also beneficial for optimizing the social applications' functionalities in terms of privacy preserving and friends recommendation. With the prevalence of SMM-oriented social apps (called SMMSA for short), this paper investigates the relationship status of SMM from a new perspective, that is, by introducing the SMM's online digital footprints left on SMMSA (e.g., presented profile, social interactions, expressions, sentiment, and mobility trajectories). Specifically, using a filtered dataset containing 2,359 active SMMSA users with their self-reported relationship status and publicly available app usage data, we explore the correlations between SMM's relationship status and their online digital footprints on SMMSA and present a set of interesting findings. Moreover, we demonstrate that by utilizing such correlations, it has the potential to construct machine-learning-based models for relationship status inference. Finally, we elaborate on the implications of our findings from the perspective of better understanding the SMM community and improving their social welfare.

Abstract P1-21-01: Factors affecting enrolment in randomised controlled trials of Japanese patients with metastatic breast cancer

Abstract Background: It is critical that a large proportion of eligible patients should participate in clinical trials. However, recruitment into clinical trials requires informed consent and involves some difficult and complex problems. Therefore, it is important to know what factors affect the participation of eligible patients in clinical trials. Patients and Methods: Patients with metastatic breast cancer (MBC) who were eligible to participate in the SELECT BC or SELECT BC-CONFIRM trials were recruited to a prospective study accompanying them named SELECT BCFEEL. The SELECT BC and SELECT BC-CONFIRM trials were randomised controlled trials (RCTs) that compared taxanes or anthracyclines, respectively, with the oral regimen of S-1 in the first line setting for Japanese patients with MBC. The SELECT BC FEEL was conducted to identify factors affecting the rates at which informed consent was obtained, using a self-administered questionnaire we developed. The questionnaire included the following items: Physician recommendation, family or friend recommendation, satisfaction with the explanation about the trial, both from the doctor or clinical research coordinator and from printed matter, regimens used, understanding of the concept of the trial, benefit to future patients, relationship with the doctor, attitude toward random assignment, concerns about privacy, burden on the patient, and potential adverse effects. We also asked the patients their age, whether they received prior chemotherapy, and about the time needed to decide whether or not to participate, communication with their doctor or other patients, potential adverse effects they were concerned about, and other people's opinions that affected their decision. The patients were asked to answer most of the questions in 5-point Likert scales. To investigate the associations between patients’ decision regarding whether or not to participate in the RCTs and their responses to each question, the Chi-square test, the Mann-Whitney U test, and the t-test were used. P values of < 0.05 were considered statistically significant. Results: Two hundred and thirty-two patients participated in the present study. Data were missing for 3 patients, and they were excluded from the analyses. One hundred and eighty five participated in the RCT, but 44 did not. The patients who chose to participate in these RCTs were more likely to answer that they decided to participate because "My doctor wanted me to participate in this trial" (69.2% vs. 25.6%; P <0.001), "My family or friends wanted me to participate in this trial" (46.7% vs. 9.3%; P< 0.001), "Both treatment regimens used in the trial are suitable for me" (48.4% vs. 23.8%; P = 0.004), "I know that the trial is conducted to determine which is a better treatment" (91.9% vs. 76.8%; P = 0.012), and "I think that my participation in the trial will contribute to the benefit of future patients with the same disease" (96.2% vs. 84.1%; P = 0.008), and spent fewer days making a decision (P =0.017) in comparison with those who declined to participate. The answering direction was reversed in only one question: those who answered "I am worried about the fact that I cannot choose which treatment to receive if I participate in the trial" were less likely to participate (46.4% vs. 67.4%; P = 0.010). The persons who had the greatest influence on the decision also differed significantly between the two groups (P = 0.005). Those who chose to participate were more likely to answer that "their doctors" were most influential (66.1% vs. 39.5%), but those who declined answered "their family" (22.8% vs. 46.5%) more frequently than the participants. Conclusions: To enhance the consent rate in randomised trials of Japanese breast cancer patients, the concepts of the trials must be considered important and acceptable by the doctors and families of the patients. Citation Format: Yasuo Hozumi, Shozo Ohsumi, Hirofumi Mukai, Masato Takahashi, Hiromitsu Akabane, Youngjin Park, Eriko Tokunaga, Tsutomu Takashima, Takanori Watanabe, Yoshiaki Sagara, Tetsuji Kaneko, Yasuo Ohashi. Factors affecting enrolment in randomised controlled trials of Japanese patients with metastatic breast cancer [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P1-21-01.

WHAT WE REMEMBER FROM SOCIAL MEDIA. MEMORY OF ADVERTISEMENTS ON SOCIAL MEDIA WEBSITES – A RESEARCH IN THE PARADIGM OF BACKWARD FRAMING

For many years human memory and its character has been the area of research interests of neurologists, psychologists and psychiatrists. Importantly, proper understanding of how our memory works is an issue that is significant from cognitive point of view and also for promotion and advertising specialists. In order to achieve the objectives of marketing communications – informational, persuasive and competitive objectives – an advertising addressee has to rightly read, remember and recall an advertisement in the phase of making the purchase decision. This process is exposed to endogenous and exogenous noises and distortions. One of the most intriguing types of memory errors is a situation, where previous experience is distorted by the information coming later on (retroactive interference). This phenomenon is being researched in the paradigm of backward framing. Works on this type of memory distortions were pioneered by Braun with co-researchers (e.g. Braun, Zaltman, 1997; Braun-LaTour, LaTour, 2005). In Poland, research on the impact of emotions on the evaluation and memory of advertisements were conducted i.a. by. A. Grochowska and A. Falkowski (2008). The source literature cites many examples of research on traditional advertising carried out in the paradigm of backward framing (i.a.. Kamins, Marks 1987; C. Hu, S. Cole, 2016, Falkowski 2017). However, there is a cognitive gap in case of this type of research being carried out in the internet environment. In case of online marketing communications, one should particularly note the promotional activities carried out on social media websites. Such communication, whose share in the market grows year by year, is characterized by certain specificity. It may be of great importance for memory and evaluation of advertisements published by that channel. In particular, they include: displaying advertisements in the form of “suggested posts”, which are additionally recommended by friends, interlarding promotional contents in memories and updates of friends displayed in news feed. The main purpose of the research being presented herein is checking how recommendations of friends and famous persons affect memory and evaluation of advertisements and how memory and evaluation of advertisements is distorted under the impact of the information coming afterwards. 102 respondents from Poland took part in the pilot research. The results indicate that there is a relationship between the contents presented on Fan Page and the subsequent product evaluation: the respondents who read comments and opinions of other internet users evaluated a non-existing product better than those who saw only an informational page. Article visualizations:

Making Explainable Friend Recommendations Based on Concept Similarity Measurements via a Knowledge Graph

Studying the similarities between the concepts in a knowledge graph can be useful in making friend recommendations on various microblogging platforms. Most existing approaches that only focus on accurate friend recommendation and can not give a reasoning explaining. In addition, existing similarity measurements are too costly and ineffective to be used in practical applications. To solve these problems, we purposed the shortest path-guide reasoning path, we perform explicit reasoning with knowledge for decision making so that the friend recommendation are supported by an interpretable causal inference. Then we designed a novel Weighted Euclidean-Shortest Path (WESP) method for measuring concept similarity in a knowledge graph and applied it to friend recommendations on a microblogging platform. First we took the shortest path as an example to measure concepts similarity. Although it was easy to use the shortest path to measure the similarity between concept pairs, the results of the measurements via the shortest path were affected by local structural imbalance in the knowledge graph. The imbalance had a significant impact on measuring concepts similarity; the more balanced the local structure, the greater similarity between the concept pairs. Then, we applied the WESP method to friend recommendations on the microblogging platform. We use the optimization similarity measurement (OSM) model that calculated the similarity between corresponding concept pairs. Our experimental results showed that the OSM method achieved better performance than the baseline methods in making friend recommendations.

Friend Recommendation Based on Multi-Social Graph Convolutional Network

Friend recommendations based on social relationships have attracted thousands of research under the rapid development of social networks. However, most of the existing friend recommendation methods use user attributes or a single social network, while rarely integrating multiple social relationships to enhance the representation. This paper focuses on integrating various social relationships to guide the representation learning, and further generating personalized friend recommendations. We design an end-to-end framework based on multiple social networks to learn the potential features of users and construct a friend recommendation model named Multi-Social Graph Convolutional Network (MSGCN). It learns the features of higher-order neighbors from multiple social networks to enrich the representation of the target user based on the improved graph convolution neural network. In particular, some graph fusion strategies by adjusting and fusing the Laplace matrix of the graph are designed to integrate social relationships. Finally, we use Bayesian theory to transform friend recommendation into a sorting problem for personalized recommendation. The experimental results show that the proposed model outperforms the state-of-the-art methods.

Hybrid Feature Based Approach for Recommending Friends in Social Networking Systems

Link prediction is an effective technique to be applied on graph-based models due to its wide range of applications. It helps to understand associations between nodes in social communities. The social networking systems use link prediction techniques to recommend new friends to their users. In this paper, we design two time efficient algorithms for finding all paths of length-2 and length-3 between every pair of vertices in a network which are further used in computation of final similarity scores in the proposed method. Further, we define a hybrid feature-based node similarity measure for link prediction that captures both local and global graph features. The designed similarity measure provides friend recommendations by traversing only paths of limited length, which causes more faster and accurate friend recommendations. Experimental results show adequate level of accuracy in friend recommendations within considerable computing time.

Privacy-Preserving Friend Recommendation in an Integrated Social Environment.

Ubiquitous Online Social Networks (OSN)s play a vital role in information creation, propagation and consumption. Given the recent multiplicity of OSNs with specially accumulated knowledge, integration partnerships are formed (without regard to privacy) to provide an enriched, integrated and personalized social experience. However, given the increasing privacy concerns and threats, it is important to develop methods that can provide collaborative capabilities while preserving user privacy. In this work, we focus on friend recommendation systems (FRS) for such partnered OSNs. We identify the various ways through which privacy leaks can occur, and propose a comprehensive solution that integrates both Differential Privacy and Secure Multi-Party Computation to provide a holistic privacy guarantee. We analyze the security of the proposed approach and evaluate the proposed solution with real data in terms of both utility and computational complexity.

Beyond safety and efficacy: sexuality-related priorities and their associations with contraceptive method selection

ObjectivesSexuality-related preferences have been understudied in contraceptive selection and uptake. Investigators endeavored to assess contraceptive preferences among patients selecting new methods at family planning clinics and to evaluate the degree to which two sexuality-related preferences are (a) valued and (b) associated with method selection. Study designData were derived from the HER Salt Lake Contraceptive Initiative, a longitudinal cohort nested in a quasi-experimental, observational study enrolling 18–45-year-old patients at four family planning health centers in Salt Lake County. At the time of selecting the new method of their choice, participants reported the importance of nine factors in contraceptive method selection, including two sexuality-related preferences: a method's lack of impact on libido and its lack of sexual interruption. Analyses involved multinomial logistic regression with method selected as the outcome, sexuality-related factors as the main explanatory variables, and a range of controls and covariates. ResultsAmong 2188 individuals seeking new contraceptive methods, the factors most frequently cited as quite or extremely important were safety (98%), effectiveness (94%), not interrupting sex (81%), not impacting libido (81%) and lack of side effects (80%). Less frequently cited factors included partner acceptability (46%), lack of hormones (39%), friend recommendation (29%) and alignment with religious beliefs (11%). Multivariate models documented no significant associations between sexual-related priorities and method selection. ConclusionsMany contraceptive seekers rank sexual-related priorities alongside safety and efficacy as very important, but a range of methods align with people's sexual priorities. ImplicationsSince patients endorse the importance of sexual-related contraceptive factors (impact on libido, impact on sexual interruption) alongside safety and efficacy, contraceptive research, counseling and care should attend to people's sexuality.

Personalized Recommendation Algorithm For Social Network Based On Two-dimensional Correlation

In order to recommend friends in a real sense based on the personalised needs of users. A personalised recommendation algorithm based on two-dimensional correlation (FRBOT) was proposed for social network. In the proposed model, the interest similarity and trust relationship among users were combined with probability matrix decomposition to analyse the potential factor characteristics of the same preferences of selected trust users and target users. Compared with general matrix decomposition algorithm and personalised recommendation method based on user trust, the algorithm has evident advantages and can improve user satisfaction. The experimental results show that the performance of the proposed friend recommendation method is significantly improved compared with that of the existing friend recommendation methods.

SIK TÜKETİLEN GIDA ÜRÜN AMBALAJ TASARIMLARININ KURUM KİMLİĞİ VE İMAJI AÇISINDAN ÖNEMİ

All of the ways an organization or organization represents its own area of work constitutes the identity of that organization or organization. Corporate identity is different from individual identity. Corporate identity refers to the identity of an organization, organization and business, similar to collective identity. The elements that make up this identity include the ways of communication of the organization, the behaviors, philosophy and visual elements of its employees. Corporate identity refers to who the organization is and what it does. Every institution needs an identity for increasing the quantity of products and for sustainable sustainable development. The ways in which institutions present their identity and the way they express themselves symbolically are becoming more and more important. In the period when tradition is at the forefront, businesses have attempted to create a corporate visual identity to show the quality of their products. Therefore, they needed designers. The identity and image of the organization includes the efforts to differentiate similar products or services from other products or services. It is therefore defined as a distinctive sign and trademark used in the provision of a particular service. Words, numbers, letters, figures, images, the shape or packaging of the product and their presentation together are considered as brands. However, it seems that in the historical process, the corporate image was initially considered only as an institution's name or a product that differentiates the product. On the other hand, the promotion and advertising of the product in the relevant media contributes to the recognition of the product than its competitors. These supports increase the appreciation and affect the purchase preferences positively. Therefore, considering that the brands without advertising and promotion opportunities are weak in terms of competition, more attention has been paid to the visual identities, designs and product quality on the shelf. Recognition of similar products offered for sale in department stores, being a priority in preferences, a good packaging with its physical structure can sell the product and give confidence to the consumer. Packaging; While supporting and increasing the quality of the product, the added value it provides and the brand identity, it reveals the differences of the product. Today, different product range continues to increase every day. Every new product that comes before the consumer can be placed on the shelves with similar products or new product types and unusual names, sometimes stored in different packages. In this case, a correctly designed packaging; the image created in the mind of the consumer. It may also aim to trigger sales, such as an individual experiment or a friend's recommendation. It can even address the heart and mind of a potential customer by providing information and training. The consumer is in the process of evaluating products; product qualities, their importance, brand beliefs, loyalty and functional value of the product. With the increase in commercial life, institutions have also been in search of identity. In order to make them accept themselves to the society and show that they are different from their competitors, institutions have started to create a separate identity for themselves. Therefore, the image of the company is all about how well this system is formed. A positive and beautiful corporate image consists of well-designed design elements and corporate identity. However, the identity and image of the corporation can be ignored and unexpected problems may arise in food product packaging designs which are mostly changed due to commercial concerns. So what the designer has to do is; improve corporate identity and image. The main purpose of this study; To emphasize the importance of frequently used food product packaging designs in terms of corporate identity and image.

Hybrid Deep Neural Networks for Friend Recommendations in Edge Computing Environment

With the rising popularity of social networks and service recommendations, research on new methods of friend recommendation have become a key topic, especially when based on quality-driven resource processing in an edge computing environment. Traditional methods seldom systematically combine static attributes (e.g., interests, geographical locations, and common friends), dynamic behaviors (e.g., liking, making comments, forwarding and @), and network structures (e.g., social ties) to recommend a new friend to a target user. Meanwhile, with the advent of deep learning, it has become more challenging to integrate these features into a deep neural network framework for friend recommendation. For example, how do we optimally make use of these features to form a united framework and what type of deep neural network architecture should be introduced into a novel recommendation method in an edge computing environment? In this paper, we propose DFRec++, a hybrid deep neural network framework combining attribute attention and network embeddings to make social friend recommendations with the help of both interactive semantics and contextual enhancement. More specifically, we first utilize the latent dirichlet allocation (LDA) topic model to generate common interest topics between users and compute the similarity of the explicit static attribute vector representation of topics, locations, and common friends. Then we feed dynamic behavior attributes into a convolutional neural network (CNN) to obtain the implicit vector representation of the interactions and context between two users. Subsequently, a multi-attention mechanism is designed to further improve the deep vector representation of the attribute information. Next, the LINE-based network embeddings algorithm is applied to embed the network structure into a low-dimensional vector. Finally, the attribute attention vector and the network embeddings are concatenated to form a deep feature representation, which is subsequently fed to a fully connected neural network (FCNN) to capture the probability of friendship between two users. The output of FCNN indicates the probability of two users becoming friends. We conducted experiments on a real-world Weibo dataset and the results show that DFRec++ outperforms several existing methods.