Semantic Scholar has informed me that my chapter with Alan Mackworth in the Handbook of Constraint Programming on Constraint Satisfaction: An Emerging Paradigm has been cited in the PREFACE to the proceedings of the International Conference on Small Islands Community: Momentum of Transformation of Small Islands Community in New Normal Era 12th December 2020, Maluku, Indonesia.

This was quite intriguing. However, it seems clear that we weren’t, in fact, cited in the preface. Perhaps in one of the papers? There are several papers in the proceedings that I would be particularly intrigued to see our connection to:

Community structure of conches (Strombus spp) in seagrass bed of Haria, Central Maluku, Indonesia

Fattening of mangrove crab (Scylla serrata) in wooden pen and plastic container

Feeding habits of giant trevally Caranx ignobilis (Carangidae) rearing in floating cage nets

Utilization of garlic as traditional fish handling in Molluccas Islands: case study on layang fish (Decapterus macrosoma, BLECKER)

Chemical properties of dried anchovy (Stolephorus sp) from Buru Island

Any of these would provide excellent further support for the claim I made in my IJCAI-20 award talk for The Ubiquity of Constraints. However, I fear Semantic Scholar has just been misled somehow.

I won’t pretend to be inundated with citations, but it is pleasant to still be getting them, and Semantic Scholar can make it easy to hear about them, by sending you emails. However, as we’ll see Semantic Scholar can sometimes use a little help 🙂 (I want to be clear though that my intention is not to diss Semantic Scholar, which is a wonderful service.)

Here are 6 citations for 6 papers that Semantic Scholar alerted me to in the past 6 days:

I received a notice that “Who ? What ? Where ? When ? Why ? and How ? 2” had cited Explaining ourselves: Human-aware constraint reasoning. Interesting title, but actually it turns out that the citing paper is titled Explainable Security. It appeared in Proceedings of the 6th Workshop on Hot Issues in Security Principles and Trust (HotSpot 2020) (another interesting title). The authors, Viganò and Magazzeni, propose “a new paradigm in security research: Explainable Security (XSec)” and discuss ‘the “Six Ws” of XSec (Who? What? Where? When? Why? and How?)’. So that explains part of the mystery of Semantic Scholar’s title for the paper, but it is still a mystery.

I was informed that Error Function Learning with Interpretable Compositional Networks for Constraint-Based Local Search cited two of my papers, In Pursuit of the Holy Grail and Progress Towards the Holy Grail. Semantic Scholar’s listing of the first paper in the References points to an entry that itself claims only 10 citations. That seemed low to me, and then I realized it was the version of the paper in ACM Computing Surveys. The version in the Constraints journal, which is the one the paper actually cites, has 59 citations. Semantic Scholar includes the second paper three times in the References, but twice in odd ways, which if hovered over declare that the paper is not yet in Semantic Scholar’s corpus.

In fact, the paper cites three of my papers; it also cites Holy Grail Redux. However, Semantic Scholar thinks Holy Grail Redux was authored by “S. Nadis” so I was not notified of this citation. Perhaps S. Nadis was and is wondering why. When you click through on Holy Grail Redux in Semantic Scholar’s list of References for this paper you arrive at a listing for Holy Grail Redux where the “Topics from this paper” include only “Bad Mojo”, and when you click on View via Publisher, you end up here!

Taking Advantage of Stable Sets of Variables in Constraint Satisfaction Problems, which I co-authored with Mike Quinn was cited by two papers. In Neural Regret-Matching for Distributed Constraint Optimization Problems, Yanchen Deng , Runshen Yu , Xinrun Wang and Bo An are “incorporating deep neural networks in solving DCOPs for the first time”. In Utility distribution matters: enabling fast belief propagation for multi-agent optimization with dense local utility function Yanchen Deng and Bo An: “theoretically show the correctness and superiority of our proposed algorithms”.

Finally, Semantic Scholar informed me that Exploring Automatic Fitness Evaluation for Evolutionary Typesetting cited a paper, Creating Personalized Documents: An Optimization Approach, which I wrote with Lisa Purvis, Steven Harrington and Barry O’Sullivan. In their paper Sérgio M. Rebelo, Tiago Martins, J. Bicker and P. Machado “present an evolutionary system for the automatic typesetting of typographic posters”. In our paper we “formalized custom document creation as a multiobjective optimization problem, and use a genetic algorithm to assemble and transform compound personalized documents”. Unfortunately, while Semantic Scholar knew enough to alert me to the citation, the one reference it lists at the citing paper’s webpage is not our paper.

I received the first notice from Semantic Scholar on June 20th and the last on June 25th, so 6 citations for 6 papers in 6 days. Just a nice coincidence, not any kind of record to be sure. Google Scholar says that Geoffrey Hinton received 85,082 citations in 2020, which works out to approximately 233 per day. Well actually “only” approximately 232, because 2020 was a leap year.

It did feel like a long year, didn’t it.

I continue to be struck by the variety of applications found in my Google or Semantic Scholar “alerts” for papers citing mine. Of course, this doesn’t necessarily mean that my work played any serious role in theirs, but since the papers of mine cited almost certainly involve constraint satisfaction it does illustrate the way constraint satisfaction can relate to so many other fields. Some recent examples:

• AdaptiFont: Increasing Individuals’ Reading Speed with a Generative Font Model and Bayesian Optimization
• Variability-intensive applications over highly-configurable platforms: Early feasibility and optimality analysis
• General Policies, Representations, and Planning Width
• Instance Space Analysis for the Job Shop Scheduling Problem
• Adaptive and interchangeable neural networks utilizing predicting information (patent)
• Confidence Adaptive Anytime Pixel-Level Recognition
• An Application of Constraint Propagation to Data-Flow Analysis