The Software Defined Network (SDN) is merged in the Internet of Things (IoT) to interconnect large and complex networks. It is used in the education system to interconnect students and teacher by heterogenous IoT devices. In this paper, the SDN-based IoT model for students’ Interaction is proposed which interconnects students to a teacher in a smart city environment. The students and teachers are free to move to anywhere, anytime and with any hardware. An architecture model for students’ teacher’s interaction in IoT is proposed which shows the details procedure about the interaction of teacher with students for electronic assessment. The SDN solves the scalability and interoperability issues between their heterogenous IoT devices. A Methodology for Students’ Answer Assessment using Latent Semantic Analysis (LSA) is proposed which calculates the semantic similarity between teacher’s question and students’ answers. The LSA is used to calculate semantic similarity between text documents. It is used to mark the students’ answers automatically by semantics. The Students’ can see results through their IoT devices just after finishing the examination with more accurate marks We have collected fifty (50) undergraduate students’ data from Learning Management System (LMS) of Virtual University (VU) of Pakistan. The experiment is implemented on eighteen (18) students’ answers in R Studio with R version 3.4.2. Teachers are provided with four (4) bins of the mark while the proposed method assigns accurate marks. The experimental results show that the proposed methodology gave accurate results as compared to teacher’s marks.

中文翻译：

---

物联网环境中基于 LSA 的 SDN 基础设施智能评估方法

软件定义网络 (SDN) 合并到物联网 (IoT) 中以互连大型复杂网络。它用于教育系统，通过异构物联网设备将学生和教师连接起来。在本文中，提出了基于 SDN 的学生交互物联网模型，该模型将学生与智慧城市环境中的教师连接起来。学生和教师可以使用任何硬件随时随地自由移动。提出了一种物联网中师生互动的架构模型，该模型显示了有关教师与学生互动进行电子评估的详细过程。SDN 解决了​​异构物联网设备之间的可扩展性和互操作性问题。提出了一种使用潜在语义分析 (LSA) 的学生答案评估方法，该方法计算教师问题和学生答案之间的语义相似度。LSA 用于计算文本文档之间的语义相似度。它用于通过语义自动标记学生的答案。学生在完成考试后可以通过他们的物联网设备查看结果，并获得更准确的分数我们从巴基斯坦虚拟大学 (VU) 的学习管理系统 (LMS) 收集了五十 (50) 名本科生的数据。该实验是在 R Studio 中使用 R 版本 3.4.2 对十八 (18) 名学生的答案进行的。向教师提供四 (4) 个分数箱，而所提出的方法会分配准确的分数。