Social Internet of Things for Domotics: a Knowledge-based Approach over LDP-CoAP

Tracking #: 1743-2955

Michele Ruta
Floriano Scioscia
Giuseppe Loseto
Filippo Gramegna
Saverio Ieva
Agnese Pinto
Eugenio Di Sciascio

Responsible editor: 
Guest Editors ST Built Environment 2017

Submission type: 
Full Paper
Ambient Intelligence aims at simplifying the interaction of a user with her surrounding context, minimizing the effort needed to increase comfort and assistance. Nevertheless, especially in built and structured environments, current technologies and market solutions are often far from providing the required levels of automation, coordination and adaptivity of the ambient. This paper proposes a novel semantic-based framework complying with the emerging Social Internet of Things paradigm. Infrastructured spaces can be intended as populated by device agents organized in social networks, interacting autonomously and sharing information, cooperating and orchestrating resources. A service-oriented architecture allows collaborative dissemination, discovery and composition of service/resource descriptions. The Semantic Web languages are adopted as knowledge representation layer and mobile-oriented implementations of non-monotonic inferences for semantic matchmaking are used to give decision capabilities to software agents. Finally, the Linked Data Platform (LDP) over the Constrained Application Protocol (CoAP) provides the knowledge organization and sharing infrastructure underpinning social object interactions. The framework has been implemented and tested in a home automation prototype integrating several communication protocols and off-the-shelf devices. Experiments advocate the effectiveness of the approach.
Full PDF Version: 

Minor Revision

Solicited Reviews:
Click to Expand/Collapse
Review #1
By Dörthe Arndt submitted on 19/Dec/2017
Minor Revision
Review Comment:

This manuscript was submitted as 'full paper' and should be reviewed along the usual dimensions for research contributions which include (1) originality, (2) significance of the results, and (3) quality of writing.

Most of the comments from my previous review are addressed in this new version of the paper. Especially the examples added helped me to understand the idea of the approach, this was a clear improvement.

The contribution of this paper, the implementation of a social internet of things framework for domotics, is explained and put into relation to other existing approaches. The authors show that their approach is able to solve the interoperability problem which often is present in AmI environments.

Nvertheless, there are still some points, I would like to see clarified. In particular:

Knowledge-based architecture (Chapter 3):
It is an improvement that you refer to Table 1 in the text and explain it later on. Nevertheless, this explanation is not close enough to the table in my opinion. I had some difficulties to find the items of the table in the text. It could for example help to use the exact names from the table in the text (e.g. use “Social agent” or “Object” in the text instead of “social object”).

Service-oriented architecture (Chapter 3):
Before you say that nodes, which are social objects, can make posts to a friend’s wall, it could be mentioned that being a social object includes the possibility of having friends.

Semantic matchmaking (Chapter 3): The examples you added really help to understand the approach, but there is still an open question:
Introducing example (starting at the end of the first column of page 5): R_1 is a subclass of S_1, so here seems to be something wrong with the example. Maybe S_1 is the class WasherDrier and R_1 is the intersection of WashingMachine and LargeCapacity?

Social entities and relationships (Chapter 3):
I have problems understanding Figure 1:
- What is ACK?
- Why does a node only get another device’s profile in the friends case (a), but not in the follower case (b), I thought that the decision to become a follower is based on the profile?
- Is the option to write on the wall reflected in case a of the figure?

Collaborative adaptivity (Chapter 3):
The concept “request” is still not clear to me: you say that “Each post contains all sensed perceptions and events observed by the social device and is considered as a request for system reconfiguration [...]” (quote from page 5). So my question here is: if this is already an observation reflecting “the world” why do I even need a process to “fulfil” the request? The process you describe seems to be some kind of bakward-reasoning which makes sure that for every observation there has at least one service that reacts to it. Is that how I need to understand it? Please clarify that in your text.

Chapter 4:
- I do not see why FullClose is selected, is the negation in Lamp_On the reason (page 12)?
- Figure 13: why does the figure contain “detectsIntrusion only IntrusionForLamp”, I see that because of the equivalence of the intrusion to the intersection of the two other intrusions that is one possible option, but “detectsIntrusion only Intrusion” would be a valid alternative in my opinion.

Minor remarks and typos:

maybe it would be helpful to better explain the problem with the current implementations and the advantage of your solution already in the abstract (you do it in the introduction, but in the abstract, you are rather vague).

Page 6:
Like in SNSs, in the framework proposed here object’s wall is the main channel for sharing knowledge. → the object’s wall

page 13: The WS selects the SC_2… → selects SC_2

page 14: As detailed in Table 3, KNX installation consisted… → the KNX installation

Review #2
Anonymous submitted on 19/Dec/2017
Major Revision
Review Comment:

I appreciate the enhancements made by the authors.

Pending Comments regarding the paper:
1- Why is the social network model used in this paper? how does it compare to other approaches based on service composition and discovery. It is not clear what is the added value of a "social model", how a "social model" is beneficial. Why such a model with a "wall", "like", "friends", "followers", and "posts" is useful. It feels like taking a "social model" and trying hard to fit in in the IoT Interoperability problem. A clarification from the authors regarding this point would be pretty appreciated.

2- The Semantic Matching is clear now with the examples. A suggestion: place the definitions in a table (R1, S1, etc). It will help the reader to quickly find the definitions and understand the concepts.

3- In Social Entities and relationships: the authors mention " .. nodes are
able to cooperate closely as they share annotations referred
to the same ontology. ". How are these annotations are set on the devices?

4- Friendship requests: the authors mention " – to decide whether to send a friendship
request to Ni
: (i) strong co-location, i.e., both devices
are placed in the same room/area; (ii) parental or coownership,
i.e., they are from the same manufacturer or
belong to the same owner; (iii) co-work, i.e., nodes are
able to cooperate closely as they share annotations referred
to the same ontology". How exactly the criteria are met? In a Smart Building hundreds of objects will be co-located. Does this mean all the devices will be friends? How does the selection and decision are made at the device level to decide between a friendship mode and a follower mode?

5- The authors mention: " weak co-work relationship,
i.e., there is low utility in a direct interaction, e.g., the
two devices are deeply different;" What is the criteria to decide that 2 devices are different? A scanner and a printer are deeply different, yet they will cooperate. When a printer or a scanner are found in the same co-location? what is expected to happen? Will they become friends or followers?

6- What will happen when applications are introduced in the Social model. Lets say a smart application which is able to read data from your phone camera and render it on the TV or send it to the printer. How is the friendship/follow approach will act in such case? Shouldn't the application decide and maintain the friendship/follow list rather than at the device level?
The reviewer cannot see how the proposed social model is distinguishing between the Applications (composing and orchestrating) between services/capabilities offered by devices and the devices themselves.

7- Collaborative adaptivity: "; it represents the percentage of coverage and
completion of the request R, as obtained by the collaborative
service discovery triggered by the post to reconfigure
the environment. ". How is the coverage percentage calculated? what is the formula? how does it relate to the Semantic Matchmaking concepts?

8- Section 3.2 depicts implementation. It is too detailed for the reader if he/she is not familiar with the technical details of CoAP. A suggestion would be to move it to the Implementation section for better clarity.

9- The authors describe in section 3.2 how their model is represented in an ontology. How dependent your model is on CoAP. How easy it can be reapplied on another protocol? It would help to dissociate the ontology from the implementation and the CoAP.

10- Section 4: Case study needs restructure. The technical details about the HEAD/GET/CoAP request should be abstracted to facilitate the flow of this section and make it more readable.

11- Use case: how did the WS decided to follow the AS? what are the conditions?

12- The Evaluation section: it would be helpful to split it into 2 sections: Implementation and Evaluation to help the reader better understand what did the evaluation cover. The section seems to show the software modules and the evaluation. It would also help to point out in subsections the criterias of evaluation.

Review #3
By Fulvio Corno submitted on 10/Jan/2018
Review Comment:

The manuscript is second submission of a paper for which a major revision was requested.
In the original comments, the soundness and originality of the work were already acknowledged by all reviewers, while some issues were raised concerning the clarity of some concepts and procedures.

In my opinion, the issues raised by the reviewers have been successfully addressed in the current revision of the paper.
The paper organization is clear and the language is fluent.