Generalized fuzzy hypergraph for link prediction and identification of influencers in dynamic social media networks

模糊逻辑 计算机科学 二元关系 鉴定(生物学) 数据挖掘 关系(数据库) 影响力营销 超图 理论计算机科学 模糊集 人工智能 数学 离散数学 植物 生物 关系营销 业务 营销 市场营销管理
作者
Narjes Firouzkouhi,Abbas Amini,Ahmed Bani‐Mustafa,Arash Mehdizadeh,Sadeq Damrah,Ahmad Gholami,Chun Cheng,Bijan Davvaz
出处
期刊:Expert Systems With Applications [Elsevier BV]
卷期号:238: 121736-121736 被引量:6
标识
DOI:10.1016/j.eswa.2023.121736
摘要

Despite the importance of link prediction and identification of influencers in dynamic social media systems, the existing methodical theories are not capable of analyzing complex multilayer relations in social media networks which contain uncertainty. In fact, there is no theoretical exploration concurrently focused on multidimensional and interrelated entities in a fuzzy-based social media environment. To cover this gap, a neoteric generalized fuzzy hypergraph (GFH) methodology is designed using developed n-ary fuzzy relation technique that is the extension of convolutional binary fuzzy relation. Characterizing reflexive, symmetric, transitive, composition, t-cut and support techniques is carried out for multidimensional uncertain-based space. Also, a graphical approach is created in the generalized fuzzy hypergraph to assist the derivation of foundational implications and concepts. The GFH framework can be applied for the intelligent management of complex systems for sole or mass users of local and global social media platforms by adopting specific membership degree for each individual. To predict the linkages between elements, a fuzzy-based indicator FLP (fuzzy link prediction) is promoted, along with the indicator of SIR (score of interaction rate) to identify the influencers (strongest communicators) in an uncertain space. Through the FLP evaluation, the extracted data are analyzed as per the highest value of 1 for single, 3 for binary, 3.8 for triplet, and 0.9 for quaternary spaces for their probable links. Through the analysis of SIR data on the individuals' membership degrees for the usage of social media platforms, the highest interaction value of 0.99 is correlated to a single member, while 5.42 magnitude addresses an influential person. The performance results show that the presented theoretical and structural approach, that is superior to the classical graph theories, is promising to configure intelligent expert systems, predict the likelihood of connections, detect communities, and specify the influencers in real social media platforms that contain uncertainty.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
1秒前
脑洞疼的应助被南乔星采纳,获得10
1秒前
2秒前
阿七发布了新的文献求助10
3秒前
彩色乐儿发布了新的文献求助10
3秒前
4秒前
RP-H完成签到,获得积分10
4秒前
5秒前
6秒前
JamesPei的应助被水天需采纳,获得10
6秒前
嘻嘻哈哈的应助被sjk采纳,获得10
6秒前
大碗宽面完成签到,获得积分10
6秒前
Orange的应助被lsq108采纳,获得10
7秒前
西红柿呀完成签到,获得积分20
8秒前
慕青的应助被cc采纳,获得10
8秒前
9秒前
去向未来发布了新的文献求助10
9秒前
9秒前
英俊的铭的应助被芊芊君子采纳,获得10
9秒前
粥末就放假完成签到,获得积分10
10秒前
10秒前
10秒前
10秒前
zeqian发布了新的文献求助10
11秒前
11秒前
zrz完成签到,获得积分10
11秒前
Seventeen完成签到,获得积分10
11秒前
onion完成签到 ,获得积分10
11秒前
笨笨芾发布了新的文献求助10
11秒前
传奇3的应助被坦率的凉面采纳,获得10
12秒前
12秒前
cdgbdfbsfdvsd完成签到,获得积分20
13秒前
13秒前
lyyu发布了新的文献求助10
13秒前
顾矜的应助被北林采纳,获得10
13秒前
蟹寒樂发布了新的文献求助10
14秒前
淳之风完成签到,获得积分10
14秒前
酱鱼发布了新的文献求助10
14秒前
合成不出来啊完成签到,获得积分10
15秒前
1234完成签到,获得积分10
16秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
八旗制度の研究 600
Improving the wear resistance of vehicle gear oil by oil soluble LaF3 nanoparticles and typical additives of gear oil 500
Boundaries and Bodhisattvas 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8183894
求助须知:如何正确求助?哪些是违规求助? 9573067
关于积分的说明 21829226
捐赠科研通 7780983
什么是DOI,文献DOI怎么找? 3434150
关于科研通互助平台的介绍 2558315
邀请新用户注册赠送积分活动 2480648