36 Matching Annotations
  1. Feb 2024
    1. https://kumu.io/

      Make sense of your messy world. Kumu makes it easy to organize complex data into relationship maps that are beautiful to look at and a pleasure to use.


      The art of mapping is to create a context in which others can think.

      Tool mentioned on [[2022-06-02]] by Jerry Michalski during [[Friends of the Link]] meeting.

  2. Sep 2023
  3. Jun 2023
    1. BookmarkNew book - Personal Knowledge Graphs, by Ivo Velitchkov


      For some additional context the work can be found through https://personalknowledgegraphs.com/#/page/pkg. It also has portions of the building of the book which exist as a knowledge graph, though it doesn't appear that they put up the entirety of the book as a linked knowledge graph the way they had initially planned. I've read a few parts in draft form, including Flancian's chapter whose ideas are tremendous, but I have yet to read the remainder of the published work.

      [Disclosure: I had submitted and had been accepted to write an early, historical-flavored chapter for this volume, but ultimately fell out, as did many others, over disagreements regarding their editing and/or publishing process. I'm close with Flancian and appreciate his experimental programming work on https://anagora.org/index, which one might call a multi-layered wiki of personal wikis, commonplace books, zettelkasten, diaries, notes, and other similar forms of personal knowledge. If you've got a public, digitally available version of a zettelkasten you'd like to add to his project, do reach out to him to interconnect it with the Agora and others' work there.]

  4. Mar 2023
  5. Jan 2023
    1. Books and Presentations Are Playlists, so let's create a NeoBook this way.


      A playlist of related index cards from a Luhmann-esque zettelkasten could be considered a playlist that comprises an article or a longer work like a book.

      Just as one can create a list of all the paths through a Choose Your Own Adventure book, one could do something similar with linked notes. Ward Cunningham has done something similar to this programmatically with the idea of a Markov monkey.

  6. Nov 2022
    1. CEO, Mike Tung was on Data science podcast. Seems to be solving problem that Google search doesn't; how seriously should you take the results that come up? What confidence do you have in their truth or falsity?

  7. Aug 2022
    1. https://www.kevinmarks.com/memex.html

      I got stuck over the weekend, so I totally missed Kevin Marks' memex demo at IndieWebCamp's Create Day, but it is an interesting little UI experiment.

      I'll always maintain that Vannevar Bush really harmed the first few generations of web development by not mentioning the word commonplace book in his conceptualization. Marks heals some of this wound by explicitly tying the idea of memex to that of the zettelkasten however. John Borthwick even mentions the idea of "networked commonplace books". [I suspect a little birdie may have nudged this perspective as catnip to grab my attention—a ruse which is highly effective.]

      Some of Kevin's conceptualization reminds me a bit of Jerry Michalski's use of The Brain which provides a specific visual branching of ideas based on the links and their positions on the page: the main idea in the center, parent ideas above it, sibling ideas to the right/left and child ideas below it. I don't think it's got the idea of incoming or outgoing links, but having a visual location on the page for incoming links (my own site has incoming ones at the bottom as comments or responses) can be valuable.

      I'm also reminded a bit of Kartik Prabhu's experiments with marginalia and webmention on his website which plays around with these ideas as well as their visual placement on the page in different methods.

      MIT MediaLab's Fold site (details) was also an interesting sort of UI experiment in this space.

      It also seems a bit reminiscent of Kevin Mark's experiments with hovercards in the past as well, which might be an interesting way to do the outgoing links part.

      Next up, I'd love to see larger branching visualizations of these sorts of things across multiple sites... Who will show us those "associative trails"?

      Another potential framing for what we're all really doing is building digital versions of Indigenous Australian's songlines across the web. Perhaps this may help realize Margo Neale and Lynne Kelly's dream for a "third archive"?

  8. Jul 2022
  9. Jun 2022
    1. https://app.thebrain.com/brains/3d80058c-14d8-5361-0b61-a061f89baf87/thoughts/32f9fc36-6963-9ee0-9b44-a89112919e29/attachments/6492d41a-73b2-20d8-b145-3283598c612b

      A fantastic example of an extensive mind map from Jerry Michalski using The Brain.

      There are lots of interesting links and resources, but on the whole

      How many of the nodes actually have specific notes, explicit ideas, annotations, or excerpts within them?

      Without these, it's an interesting map and provides some broad context, but removes local specific context of who Jerry is and how he explicitly thinks. One can review the overarching parts to extract what his biases may be based on availability heuristics, but in areas of conflicting ideas which have relatively equal numbers of links within a particular area, one may not be able to discern arguments from each other.

      Still a fascinating start and something not commonly seen in the broader literature.

      I'll also note that even in a small sample of one video call with Jerry sharing his screen while we talked about a broad sub-topic it's interesting to see his prior contexts as we conversed. I've only ever had similar experiences with Bill Seitz who regularly drops links to his wiki pages in this sort of way or Kevin Marks (usually in text chat contexts and less frequently in video calls/conversations) who drops links to his extensive blogging history which also serves to add his prior thoughts and contextualizations.

  10. May 2022
    1. https://www.otherlife.co/pkm/

      The PKM space has gotten crazy, but mostly through bad practice, lack of history, and hype. There are a few valid points I see mirrored here, but on the whole this piece is broadly off base due to a lack of proper experience, practice and study. I definitely would recommend he take a paid course to fix the issue, but delve more deeply into recommended historical practices.

    2. The single most widely shared marketing image for Roam Research

      This useless knowledge graph is one of the worst parts about Roam Research. It is bad UI and wholly unusable.

  11. Apr 2022
    1. hey can be aesthetic, perhaps. In which case, that corroborates my theory. You’re not accumulating knowledge and insight—you’re drawing pretty pictures.

      I think there's definitely an aesthetic enjoyment factor, but that doesn't mean it's bad -- if anything it might be the opposite?

      You could imagine describing a knowledge graph as 'pleasant' -- or even 'beautiful'. You can see them as "pretty pictures" -- or you can see them as (generative?) art.

  12. Feb 2022
    1. Dafür wird der Wissensgraph um geeignete Tools erweitert. DasTechnologiespektrum reicht hier je nach Strukturiertheitsgrad der Daten von Methodender semantischen Textanalyse (vgl. [6]) über das Parsen von regulären Ausdrücken(s. Abschn. 6.4.2) bis hin zum (teil)automatischen Mappen mithilfe von Transforma-tionsvokabularen (z. B. D2RQ in [7], R2RML)

      Beispiel für eine KG-Erweiterung

    1. Verbesserungspotenzial im Bereich der Graph-Visualisierung, des Authorings von Ontologien und Regeln und der einfachen Anbindung weiterer Datenquellen.

      Verbesserungspotential von EKGs

    2. Für eine noch schnellere Verbreitung im Unternehmensumfeld müssen die zugrunde liegenden Technologien jedoch zugänglicher für Nicht-Techniker werden.

      Zukunft: Zugang für Nicht-Techniker

    3. Projektgraphen um die Fähigkeit erweitert, dem Projektleiter basierend auf seinen beschreibenden Texten relevante Themen und Technologien zur Übernahme vorzuschlagen.

      Einsatz von Projektgraphen Erweiterung um die Fähigkeit, basierend auf seine beschreibenden Texten relevante Themen und Technologien zur Übernahme vorzuschlagen

    1. Der im Projekt „Smart Data Web“ erstellte öffentliche Teil des Wissensgraphen wurde zudem zum Aufbau eines Siemens-internen Corporate Knowledge Graphen genutzt. Dazu wurden relevante Teilmengen des öffentlichen Wissensgraphen extrahiert und in das ge-schützte Siemens- Netzwerk transferiert. Die internen Datenbanken von Siemens wurden nach RDF konvertiert und zusammen mit dem SDW KG in eine geschützte Datenbank geladen. Weiterhin wurden vom Anwendungsfall getriebene Abfragen erstellt, welche in-terne und offene Daten kombinieren. Der Corporate Knowledge Graph (CKG) ermöglicht eine einheitliche, konsistente und elegante Verknüpfung interner und externer Informatio-nen, ganz im Sinne einer „Enterprise-Intelligence“-Lösung. Über den CKG können Infor-mationen, im konkreten Fall zu Zulieferern, aggregiert und konsolidiert abgerufen und für die Einkaufsabteilungen von Siemens dargestellt werden. Dabei werden interne Kennzah-len, z. B. zum Projektvolumen und zu Bewertungen einzelner Lieferanten, mit aktuellen, automatisch gesammelten, firmen-, produkt- und standortbezogenen Ereignissen aus Nachrichten und anderen Textdatenquellen verknüpft, sodass die Anwender eine Gesamt-sicht auf entscheidungsrelevantes Wissen erhalten

      Projekt „Smart Data Web“ Corporate Knowledge Graph (CKG) - ermöglicht eine einheitliche, konsistente und elegante Verknüpfung interner und externer Informationen, ganz im Sinne einer „Enterprise-Intelligence“-Lösung

      Semantische Verknüpfung/Ontologie:

      Dabei werden interne Kennzah- len, z. B. zum Projektvolumen und zu Bewertungen einzelner Lieferanten, mit aktuellen, automatisch gesammelten, firmen-, produkt- und standortbezogenen Ereignissen aus<br /> Nachrichten und anderen Textdatenquellen verknüpft

      Potential: eine Gesamt- sicht auf entscheidungsrelevantes Wissen erhalten.

    1. Enterprise Knowledge Graphs (EKGs) mightbe considered as an embodiment of LED

      Enterprise Knowledge Graphs (EKGs) als eine Verkörperung von LED

    2. Sören Auer
    3. The unified approachhas the advantage, that the enterprise has more control overthe data and quality, and the data querying is significantlyfaster.
    4. REFERENCES[1] C. Bizer, J. Lehmann, G. Kobilarov, S. Auer, C. Becker, R. Cyganiak,and S. Hellmann. Dbpedia-a crystallization point for the web of data.Web Semantics: science, services and agents on the world wide web,7(3):154–165, 2009.[2] D. Calvanese, M. Giese, D. Hovland, and M. Rezk. Ontology-basedintegration of cross-linked datasets. In Proceedings of the 14th Interna-tional Semantic Web Conference (ISWC). Springer, 2015.[3] X. Dong, E. Gabrilovich, G. Heitz, and W. Horn. Knowledge vault: Aweb-scale approach to probabilistic knowledge fusion. In Proceedingsof the 20th ACM SIGKDD international conference on Knowledgediscovery and data mining, pages 601–610, 2014.[4] P. Frischmuth, S. Auer, S. Tramp, J. Unbehauen, K. Holzweißig,and C. Marquardt. Towards linked data based enterprise informationintegration. In S. Coppens, K. Hammar, M. Knuth, and et al., editors,Proceedings of the Workshop on Semantic Web Enterprise Adoption andBest Practice (ISWC 2013), 2013. CEUR-WS.org, 2013.[5] R. Isele and C. Bizer. Active learning of expressive linkage rules usinggenetic programming. Web Semantics: Science, Services and Agents onthe World Wide Web, 23:2–15, 2013.[6] L. Masuch. Enterprise knowledge graph - one graph to connect themall. 2014.[7] P. N. Mendes, H. Mühleisen, and C. Bizer. Sieve: Linked data qualityassessment and fusion. In Proceedings of the 2012 Joint EDBT/ICDTWorkshops, pages 116–123, 2012.[8] J. Michelfeit, T. Knap, and M. Neˇcask `y. Linked data integration withconflicts. arXiv preprint arXiv:1410.7990, 2014.[9] A.-C. Ngonga Ngomo and S. Auer. Limes - a time-efficient approachfor large-scale link discovery on the web of data. In Proceedings ofIJCAI, 2011.[10] N. F. Noy. Semantic integration: a survey of ontology-based approaches.ACM Sigmod Record, 33(4):65–70, 2004.[11] T. Pellegrini, H. Sack, and S. Auer, editors. Linked Enterprise Data.X.media.press. Springer, 2014.[12] A. Schultz, A. Matteini, R. Isele, P. N. Mendes, C. Bizer, and C. Becker.Ldif-a framework for large-scale linked data integration. In 21stInternational World Wide Web Conference (WWW 2012), DevelopersTrack, Lyon, France, 2012.
    5. In general, a federated approach will be advan-tageous if the enterprise aims to continuously ingest updatesand new additions from public LOD sources.
    6. Nevertheless, acertain overhead for query expansion and entailment regimesis required.
    7. Enterprise Knowledge Graphs

      The unified approach has the advantage, that the enterprise has more control over the data and quality, and the data querying is significantly faster.

    8. Enterprise Knowledge Graphs are the next stage in theevolution of knowledge management systems.

      Enterprise Knowledge Graphs are the next stage in the evolution of knowledge management systems.

    1. Knowledge graph (KG) embedding is to embed components of a KG including entities and relations into continuous vector spaces, so as to simplify the manipulation while preserving the inherent structure of the KG. It can benefit a variety of downstream tasks such as KG completion and relation extraction, and hence has quickly gained massive attention. In this article, we provide a systematic review of existing techniques, including not only the state-of-the-arts but also those with latest trends. Particularly, we make the review based on the type of information used in the embedding task. Techniques that conduct embedding using only facts observed in the KG are first introduced. We describe the overall framework, specific model design, typical training procedures, as well as pros and cons of such techniques. After that, we discuss techniques that further incorporate additional information besides facts. We focus specifically on the use of entity types, relation paths, textual descriptions, and logical rules. Finally, we briefly introduce how KG embedding can be applied to and benefit a wide variety of downstream tasks such as KG completion, relation extraction, question answering, and so forth.

      Bei der Einbettung von Wissensgraphen (KG) werden die Komponenten eines KG, einschließlich Entitäten und Beziehungen, in kontinuierliche Vektorräume eingebettet, um die Bearbeitung zu vereinfachen und gleichzeitig die inhärente Struktur des KG zu erhalten. Sie kann für eine Vielzahl von nachgelagerten Aufgaben wie KG-Vervollständigung und Relationsextraktion von Nutzen sein und hat daher schnell große Aufmerksamkeit erlangt. In diesem Artikel geben wir einen systematischen Überblick über die vorhandenen Techniken, wobei wir nicht nur den aktuellen Stand der Technik, sondern auch die neuesten Trends berücksichtigen. Dabei wird insbesondere auf die Art der bei der Einbettung verwendeten Informationen eingegangen. Zunächst werden Techniken vorgestellt, die die Einbettung nur anhand der in der KG beobachteten Fakten durchführen. Wir beschreiben den allgemeinen Rahmen, das spezifische Modelldesign, typische Trainingsverfahren sowie die Vor- und Nachteile solcher Techniken. Danach werden Techniken diskutiert, die neben Fakten auch zusätzliche Informationen einbeziehen. Wir konzentrieren uns insbesondere auf die Verwendung von Entitätstypen, Beziehungspfaden, textuellen Beschreibungen und logischen Regeln. Abschließend stellen wir kurz vor, wie die KG-Einbettung auf eine Vielzahl von nachgelagerten Aufgaben wie KG-Vervollständigung, Beziehungsextraktion, Beantwortung von Fragen usw. angewendet werden kann und davon profitiert.

    1. Practical guidance on KR, knowledge graphs, semantic technologies, and KBpedia

      Titel: A knowledge representation practionary Autor: Michael K. Bergman

  13. Dec 2020
    1. introduction and in its summary at the end: “a graph of data intended to accumulate and convey knowledge of the real world, whose nodes represent entities of interest and whose edges represent relations between these entities”. We

      comprehensive definition

    2. but these same vendors were also talking about Semantic Web and Linked Data capabilities before that, so I thought that they were just rebranding with the new buzz phrase as a marketing strategy.

      Maybe the vendor dependency is one of the problems.

  14. Apr 2019
  15. Nov 2018
    1. Creating KGs is not trivial.

      This applies to universal KG in particular. Domain specific KGs can have any level of complexity - can they still be called knowledge graphs then?