Table 4 Experimental datasets used in the present novel ML models after considering outlier’s removal.

From: An efficient machine learning approach for predicting concrete chloride resistance using a comprehensive dataset

Reference	Count	W/B (–)	Cement (kg/m³)	Slag (kg/m³)	FA (kg/m³)	SF (kg/m³)	Fine aggregate (kg/m³)	Coarse aggregate (kg/m³)	SP (% by binder wt.)	Fresh density (kg/m³)	\({f}_{c}\) (MPa)	\({D}_{nssm}\)
Costa & Appleton (1999)⁶⁴	24	0.30–0.50	300–500	0	0	0–21.5	0–822	1131–1704	0–0.02	×	34–66	0.88–4.83
Thomas & Bamforth (1999)⁶⁵	18	0.48–0.66	110–288	0–255	0–98	0	585–660	1240–1305	0	×	37.9–49.6	0.59–10
Hao–bo & Guo–zhi (2004)⁶⁶	16	0.36	192–480	0–147	0–216	0–24	614–696	1061–1147	0.9–1.0	×	49.1–63.1	0.74–7.0
Alizadeh et al. (2008)⁶⁷	2	0.5	370–400	0	0	0–30	778	956	0–0.01	2355–2370	39–54.1	0.99–1.51
Song & Kwon (2009)³¹	120	0.37–0.47	178–454	0–227	0–136	0–23	745–838	921–976	0.01	×	×	0.22–6.3
Shekarchi et al. (2009)⁶⁸	60	0.35–0.50	350–400	0	0	0–50	778–936	955–1022	0–0.02	2255–2467	33.9–79.6	0.22–12.54
Audenaert & De Schutter (2010)⁶⁹	20	0.37–0.60	300–450	0	0–240	0	626–923	583–1225	0	×	40.3–74.7	0.15–13.5
Jain & Neithalath (2011)⁷⁰	15	0.40	344–430	0	0–86	0–39	730–743	1050–1058	0	×	×	1.65–9.94
Liu et al. (2011)⁷¹	21	0.38–0.54	400–500	0	0	0	321–517	518–1098	0–0.65	1610–2360	31–75	6.5–19.1
Maes et al. (2013)⁷²	4	0.50	52–350	0–295	0	0	785–791	1036–1043	0	2331–2350	×	3.93–8.98
Elfmarkova et al. (2015)⁷³	4	0.40	312.3–520.6	0–312.3	0–442.5	0–468.5	1574.1	0	0.2–0.4	2150–2210	51.9–60.4	2.49–11.15
Real et al. (2015)⁷⁴	86	0.35–0.55	175–450	0	0–180	0–45	703–1272.7	0–593.2	0	×	16.9–84.2	3.8–22.8
Bogas & Gomes (2015)⁷⁵	49	0.30–0.55	270–525	0	0–735	0–36	646–1073	266.35–915.53	0–0.83	1483–2411	30.9–76.2	3.4–19.9
Liu et al. (2015)⁷⁶	16	0.30–0.54	144–471	0	0	0	235–874	255–1064	0–4.17	1364–2365	×	4.9–19.1
Farahani et al. (2015)⁷⁷	12	0.35–0.5	350–370	0	0	30–50	782–931	955–1020	0–0.02	×	×	0.33–2.7
Park et al. (2016)⁷⁸	105	0.35–0.55	128–470	0–282	0–141	0	669–815	927–978	0.7–1.22	×	×	1.02–42.04
Ferreira et al. (2016)⁷⁹	15	0.45–0.60	330–440	0	0	0	600–710	1240–1240	0	×	18.9–44.8	5.7–54
Pilvar et al. (2016)⁸⁰	24	0.35–0.45	297.5–400	0	0	0–60	1037–1150	692–766	0–1.10	×	46–80	1.41–14.3
Choi et al. (2017)⁸¹	12	0.40–0.60	350	0	0	0	707–742	1080–1135	0	×	×	8.74–133.6
Liu et al. (2017)⁸²	15	0.38–0.53	286–454	0	0–123	0	689–729	1041–1094	0	×	34.6–74.3	2.3–9.27
Shiu & Yang (2020)⁸³	16	0.34– 0.65	168–495	0– 242	0	0	786	934	0– 2.07	×	231– 483	3.24– 26.7
Naito et al. (2020)⁸⁴	138	0.19–0.65	13.02–2384.97	0–1284.4	0–74.2	0–29.7	27.5–904.9	26.4–1187.2	0	2176.9–2447.6	19.2–80.6	0.55–30.6
Sell Junior et al. (2021)⁸⁵	6	0.45–0.65	320–400	0	0	80	587.2–640.0	880.8– 960.0	0	×	40.6– 74.5	0.20– 6.02
Pontes et al. (2021)⁸⁶	20	0.35– 0.55	245–450	0	0– 135	0– 27	606.2– 822.2	935.1– 1154.96	0	×	26.1– 78.0	5.9– 25.2

Back to article page

Table 4 Experimental datasets used in the present novel ML models after considering outlier’s removal.

Search

Quick links