Service recommendation is widely used to locate developers' desired services. Previous methods mainly focus on employing collaborative filtering (CF) techniques recommend services developers. However, these have some problems, such as being sensitive sparse data and having limited predictive ability new Generative adversarial network (GAN) can solve the above mentioned since it learn distribution from a amount of generate developer's preference score for service, even if he/she has not...

Abstract Nowadays, collaborative filtering recommender systems have been widely deployed in many commercial companies to make profit. Neighborhood‐based (CF) is common and effective. To date, despite its effectiveness, there has little effort explore their robustness the impact of data poisoning attacks on performance. Can neighborhood‐based be easily fooled? this end, we shed light propose a novel attack framework, encoding purpose constraint against them. We first illustrate how calculate...

The sequential recommendation has been widely used to predict users' preferences in the near future by utilizing their dynamic interactions with items. However, existing methods only consider single-typed (e.g., purchase), ignoring rich heterogeneous information such as multi-typed click, purchase) and item attributes (e.g, category), which leads a suboptimal model. We can integrate this introducing Dynamic Heterogeneous Information Networks (DHINs). Our solution contains three special...

With the popularity of smartphones, mobile applications (mobile apps) have become a necessity in people's lives and work. Massive apps provide users with variety choices, but also bring about information overload problem. In reality, number that used is very limited, resulting sparse interaction matrix between apps. It not accurate enough to use predict numerous unknown ratings, so recommended results cannot satisfy users. This paper aims exploit user's historical behavior data app's side...

In this paper, multi-switching period oscillation phenomenon in constant on-time (COT) controlled buck converter is studied, and the effect of output capacitor equivalent series resistance (ESR) on control performance COT revealed. The study results indicate that ESR critical factor causing converter, value obtained. When less than value, occurs, it disappears when higher value. Finally, theoretical are verified by simulation.

Ranking items to users is a typical recommendation task, which evaluates users' preferences for certain over others. Easy access social networks has motivated researchers incorporating trust information recommendation. In this paper, aiming at offering fundamental support the trust-based research item recommendation, we conduct an in-depth analysis on Epinions, Ciao, and FilmTrust data sets. We find that user's selection of influenced not only by her trustees but also trusters. leverage...

Distrust based recommender systems have drawn much more attention and became widely acceptable in recent years. Previous works investigated using trust information to establish better models for rating prediction, but there is a lack of methods distrust relations derive accurate ranking-based models. In this article, we develop novel model, named TNDBPR (Trust Neutral Bayesian Personalized Ranking), which simultaneously leverages trust, distrust, neutral item ranking. The experimental...

Nowadays, collaborative filtering recommender systems have been widely deployed in many commercial companies to make profit. Neighbourhood-based is common and effective. To date, despite its effectiveness, there has little effort explore their robustness the impact of data poisoning attacks on performance. Can neighbourhood-based be easily fooled? this end, we shed light propose a novel attack framework encoding purpose constraint against them. We firstly illustrate how calculate optimal...

Abstract Background: To evaluate the effects of total hip replacement and internal fixation on outcome femoral neck fractures in elderly. Methods: A retrospective study was conducted a 79 individuals diagnosed with Garden I or II aged from 56-71 years old October 2012 to May 2019. Patients treated were grouped into group 1, while those 2. Baseline characteristics compared between two groups eliminate extra factors. Postoperative activity time, blood loss, length hospital stay, surgical...

Recommending suitable services to users autonomously has become the key solve problem of service information overload. Existing recommendation algorithms have some limitations, and discard side node, or ignore intermediate omit feature neighbouring nodes, not model pairwise attentive interactions between services. To above-mentioned this paper proposes a approach by leveraging graph attention network (GAT) co-attention mechanism in heterogeneous networks (HINs). Specifically, different types...