InoteveraGaidhi rinotevera
Rangarira, Precision uye Elusion Kuedzwa mune Gwaro Ongororo
Tekinoroji
Basics GUIDE
Precision uye Recall
Kurongeka uye kuyeuka ndiwo maviri anoenderana metrics ekuongorora classifiers, kunyanya kana makirasi asina kuenzana.
Teerera kugwaro iriAI-yakarondedzerwa · ~2 min
Pa peji ino2 min verenga
Pfupiso
Pamwe chete vanoratidza izvo zviri pachena zvinovanza - kangani fungidziro yakanaka yemuenzaniso ndeyechokwadi, uye kuti mangani echokwadi echokwadi aanobata.
Kudzika Kwakadzika
Kana modhi ichiratidzira zvinhu sezvakanaka, mibvunzo miviri inokosha. Precision inobvunza: pane zvese zvatakamaka, zvingani zvaive zvakanaka zvechokwadi? Izvo zvakaenzana nezvakanaka zvechokwadi zvakakamurwa nezvose zvakafanotaurwa zvakanaka, kuranga maaramu enhema. Rangarira (senitivity) inobvunza: pane zvese zvakanaka zvakanaka kunze uko, vangani vatakabata? Inoenzana nemaposisiti echokwadi akapatsanurwa neese mapositisti chaiwo, kurangwa kwekupotsa. Izvi zvinowanzochinjana: kudzikisa chikumbaridzo chesarudzo kunobata zvimwe zvakanaka (yepamusoro kuyeuka) asi kunoisa mireza yakawanda (yakaderera chaiyo), uye zvichipesana. Izvo zvekuisa pamberi zvinoenderana nemitengo - spam sefa inofarira chaizvo (usarase tsamba chaiyo), nepo kenza skrini ichifarira kuyeuka (usapotsa bundu). Iyo F1 mamakisi, yavo harmonic zvinoreva, zviyero zvese munhamba imwe.
Kuitwa muNyika yeChokwadi
Masefa eSpam anoteedzera zvakanyanya saka maemail ari pamutemo anenge asina kumbotumirwa zvisizvo kufolda yespam.
Miedzo yekuongorora yekurapa inotungamira kuyeuka kwakanyanya kudzivirira varwere vanoshayikwa vane chirwere ichi, vachigamuchira mamwe manyepo ekutevera.
Kutsvaga uye kurudziro masisitimu anoshuma precision@k (mangani epamusoro k mhinduro akakodzera) kuyera nhanho yemhando.
Kuonekwa kwehutsotsi kunoenzanisa nemazvo uye kuyeuka kuburikidza neF1 chibodzwa, sezvo maaramu enhema nehutsotsi hwakarasika zvinodhura.
Ramangwana Rokunyatsojeka uye Kurangarira
Sezvo AI ichipinda munzvimbo dzepamusoro-soro - kuongororwa kwekurapa, kurongeka kwemukati, kubiridzira - zvikwata zvinowedzera kurondedzera nemazvo uye kuyeuka (uye macurves avo) pane kurongeka kwega, uye tune zvikumbaridzo kuti zvienderane nemitengo yepasirese uye zvipingaidzo zvekururamisira. Per-boka nemazvo/kurangarira odhita ari kuita chiyero chekuona akasiyana mhosho pahuwandu hwevanhu. Tarisira akapfuma-asingatarisire mari metrics, akarongedzerwa mikana, uye midziyo inoita kuti vanobatana vasarudze mapoinzi ekushanda vachipindirana pane kubvuma chikumbaridzo che 0.5.
Kuzvishandisa
Nei zvichikosha
- Sarudzo dzakajeka. Inokubatsira kuparadzanisa zvakajeka zvichemo zvehunyanzvi kubva mumutauro wekushambadzira.
- Mutengo uye bhajeti. Iwe unogona kubvunza zvirinani kuita mibvunzo usati washandisa mari kana nguva.
- Team uye workflow. Zvikwata zvine nzwisiso yakagovaniswa inoita zvirinani chigadzirwa, mutemo, uye sarudzo dzekudzidza.
Chenjera
- Zvikwata zvakasiyana zvinogona kushandisa izwi rimwechete zvakasiyana, saka tsanangura nzvimbo nekukurumidza.
- Benchmarks inogona kutaridzika yakasimba nepo chaiyo-yenyika kuita isina kuenzana.
- Kuregeredza mhando yedata uye zvirongwa zvekuongorora zvinowanzogadzira mhedzisiro isina kusimba.
Matanho anotevera
- Tanga netsanangudzo yemutauro wakajeka yemhedzisiro yaunoda.
- Sarudza metric imwe yekubudirira uye imwe yekutadza mamiriro usati waedzwa.
- Mhanya mutyairi mudiki ane data remumiriri, kwete demo rakakwenenzverwa.
- Nyora uko Precision uye Recall inobatsira uye uko nzira dzakareruka dziri nani.
Ramba Uchiongorora
Tsamba yemahara
Tora iyo yemazuva ese AI pfupiso
Nhatu dzakasimbiswa dzeAI nyaya mangwanani ega ega pakati pevhiki, dzakanyorwa neChirungu chakajeka. Yemahara zvachose, hapana ads.
Imwe email zuva rega rega revhiki. Bvisa kunyoresa kamwe chete. Hatimbotengese kana kugovera kero yako.
Zviedze iwe pachako
Take the Precision and Recall quiz
Mhinduro yepakarepo pamhinduro yega yega, uye chitupa chinogovaniswa chine ID inobatika kana wapedza kosi.
Tsigira dzidzo yemahara yeAI. AI Understanding ndeye 501(c)(3) isingabatsiri - hapana ads, hapana paywall, zvachose. Ita chipo
Mibvunzo inowanzo bvunzwa
Chii chinonzi Precision uye Recall?
Kurongeka uye kuyeuka ndiwo maviri anoenderana metrics ekuongorora classifiers, kunyanya kana makirasi asina kuenzana. Pamwe chete vanoratidza izvo zviri pachena zvinovanza - kangani fungidziro yakanaka yemuenzaniso ndeyechokwadi, uye mangani echokwadi echokwadi aanobata.
Precision inopindura mubvunzo upi?
Precision = TP / (TP + FP) inoyera kurongeka kwekufanotaura kwakanaka kwemuenzaniso - kuvimbika kwemureza wakanaka.
On a fixed evaluation set, lowering a positive-class decision threshold usually produces which tradeoff?
Lowering the threshold expands or preserves the predicted-positive set. It may catch additional true positives and additional false positives; changes depend on which items cross the threshold.
Why can precision and recall reveal a weakness hidden by high accuracy on a dataset dominated by negatives?
Accuracy can be dominated by correct negative predictions. Precision and recall examine positive flags and missed positives using TP, FP, and FN.
Six of the top ten search results are relevant. What is precision@10 for that result list?
Precision@10 measures the relevant fraction among the top ten results: 6/10 = 60%.
Ramba uchidzidza
Mazano anoenderana
Mamwe madhairekitori akasarudzirwa nyaya iyi