- From: Harshvardhan J. Pandit <me@harshp.com>
- Date: Thu, 21 Dec 2023 12:24:12 +0000
- To: Arthit Suriyawongkul <arthit@gmail.com>, Data Privacy Vocabularies and Controls Community Group <public-dpvcg@w3.org>
Hi Art. Thanks for the proposals. What is the motivation here to have these concepts? And why select only these two specific concepts rather than the entire taxonomy from IAB/Google? The IAB Audience and Google Taxonomies for data are something we have discussed - but there are issues with adopting them as-is such as identifying sensitive/special categories. If they are (also) to be adopted as-is, I had proposed they be modelled separately as an extension over DPV-PD to keep the external taxonomy as 'external' within DPV. Regarding the concepts: 1) AdvertisingID should be further specialised as DeviceAdvertisingID if it is only for the device, and should be a subclass of Identifier 2) PurchaseIntention should be PurchaseIntent I think (for phrasing) Regards, Harsh On 20/12/2023 22:23, Arthit Suriyawongkul wrote: > Classes related to advertising industry: > > (1) > IRI https://w3id.org/dpv/dpv-pd#AdvertisingID > Term: AdvertisingID > Label: Advertising ID > Description: Information about a unique user ID assigned to a device > or an operating environment, to help advertising services personalize > their offers. > SubType of: dpv-pd:DeviceBased dpv-pd:Identifying > Source: https://en.wikipedia.org/wiki/Advertising_ID and audience_id > in IAB Tech Lab Data Transparency Standard 1.1 > https://iabtechlab.com/standards/data-transparency/ > > (2) > IRI https://w3id.org/dpv/dpv-pd#PurchaseIntention > Term: PurchaseIntention > Label: Purchase Intention > Description: Information about current in-market purchase intent > SubType of: dpv-pd:Intention dpv-pd:Transactional > Source: IAB Tech Lab Audience Taxonomy > https://iabtechlab.com/standards/audience-taxonomy/ > See Also: dpv-pd:Purchase > > -- > > "An advertising ID is a unique user ID assigned to a mobile device > (smart phone, tablet computer), or operating environment, to help > advertising services personalize their offers.[1] It can be sent to > advertisers and other third parties which can use this unique ID to > track the user's movements, habits, and usages of applications.[2] > There is a potential for such technology to replace magic cookies." > > https://en.wikipedia.org/wiki/Advertising_ID > > -- > > In its Audience Taxonomy, > Interactive Advertising Bureau (IAB), uses demographic, purchase > intent, and interest for the audience segment. > > DPV-PD already has dpv-pd:Demographic and dpv-pd:Interest classes. > Having a dpv-pd:PurchaseIntention will make DPV-PD able to closer > align with practices in the advertising industry. > > *I use the term (Purchase) "Intention" here instead of "Intent", as > appeared in IAB Audience Taxonomy, to make it consistent with > dpv-pd:Intention. > > While DPV already has dpv-pd:Purchase and > dpv-pd:PurchasesAndSpendingHabit classes, > they are both for purchases that have been made in the past. > > -- > > For an illustration of how these Demographic, Interest, and Purchase > Intention are used, consider a possible taxonomy_id_list, > which is a required field in the IAB Tech Lab Data Transparency > Standard 1.1 Minimum Segment Disclosure. > > "7;20;66;123;246;641;843|PIPR8;824|PIFI2" > > - ID 7 : Demographic | Age Range | 35-39 | > - ID 20 : Demographic | Education & Occupation | College Education | > - ID 66 : Demographic | Household Data | $50000 - $74999 | > - ID 123: Demographic | Household Data | 2 Adults | > > - ID 246: Interest | Automotive | Auto Technology | > - ID 641: Interest | Sports | Field Hockey | > > - ID 824: Purchase Intent* | Automotive Ownership | Green Vehicles | > - has New Vehicles as a Parent > - Future Buyer Intent Medium > > - ID 843: Purchase Intent* | Automotive Ownership | Budget Cars | > - has Pre-Owned Vehicles as a Parent > - with a modifier PIPR6: Past Purchase Recency > 12 months > > The example is adapted from these documents > [1] https://iabtechlab.com/wp-content/uploads/2022/08/IAB-TL-Data-Transparency-Standard-Disclosure-Schema-1.1-Final-1.26.21.pdf > [2] https://iabtechlab.com/wp-content/uploads/2022/02/IAB-Tech-Lab-Taxonomy-and-Data-Transparency-Standards-to-Support-Seller-defined-Audience-and-Context-Signaling.pdf > [3] https://app-api.datalabel.org/docs/enums > > -- > > To be more conservative and not introducing unnecessary new classes: > > When look at Purchase Intent modifiers in [3], > - a user can declare their own PIPRs (Past Purchase Recency) and PIPFs > (Past Purchase Frequency) can be subtypes of dpv-pd:Purchase, > - a user can also declare their own PIFIs (Future Buyer Intent) can be > a subtype of dpv-pd:PurchasesAndSpendingHabit - if the "Habit" can > also include the tendency in the future. > > And maybe there will be no need to have a separate dpv-pd:PurchaseIntention. > > -- > > cheers, > Art > -- --- Harshvardhan J. Pandit, Ph.D Assistant Professor ADAPT Centre, Dublin City University https://harshp.com/
Received on Thursday, 21 December 2023 12:24:22 UTC