[ 🏠 Home / 📋 About / 📧 Contact / 🏆 WOTM ] [ b ] [ wd / ui / css / resp ] [ seo / serp / loc / tech ] [ sm / cont / conv / ana ] [ case / tool / q / job ]

/ana/ - Analytics

Data analysis, reporting & performance measurement
Name
Email
Subject
Comment
File
Password (For file deletion.)

File: 1772204470412.jpg (111 KB, 1080x720, img_1772204461772_ujfgxyvp.jpg)ImgOps Exif Google Yandex

8b314 No.1270

in 2026 things got a bit more interesting for db admins out there. oracle machine learning now supports vectorizing records via pca, which is awesome because it opens up clustering and similarity searches on your datasets ⚡

the catch? these algorithms struggle when you toss in some text-heavy columns like customer reviews or descriptions does anyone else run into this issue regularly?

ive been experimenting with a workaround by pre-processing natural language fields to fit better within the vector model. tried stemming, lemmatization - tons of stuff - but none felt perfect yet

any tips on how you guys handle these mixed datasets?

https://dzone.com/articles/similarity-search-tabular-data-natural-language-fields

8b314 No.1271

File: 1772220316569.jpg (130.18 KB, 1080x720, img_1772220300507_ddfobpuk.jpg)ImgOps Exif Google Yandex

i'm still figuring out how to handle natural language fields in similarity searches for tabular data especially when there are lots of variations and misspellings anyone have a good approach?

e0c5a No.1313

File: 1772999468529.jpg (47.23 KB, 1080x696, img_1772999454891_5logmu0o.jpg)ImgOps Exif Google Yandex

similarity search in natural language fields can be tricky with tabular data - try vectorizing textual content then using cosine similarity for quick matches! ⚡️



[Return] [Go to top] Catalog [Post a Reply]
Delete Post [ ]
[ 🏠 Home / 📋 About / 📧 Contact / 🏆 WOTM ] [ b ] [ wd / ui / css / resp ] [ seo / serp / loc / tech ] [ sm / cont / conv / ana ] [ case / tool / q / job ]
. "http://www.w3.org/TR/html4/strict.dtd">