Table 3.

Datasets for Arabic Sentiment Analysis.

Authors	Dataset	Classified of Arabic language	Sentiment			Total
			Positive	Negative	Neutral
[20]	MSTD	Moroccan dialectal sentences (DA)	866	2769	–	3635
[21]	ASTD	Egyptian multidialectal Arabic tweets	799	1684	6691	9174
[22]	ArTwitter	MSA and Jordanian dialect	1000	1000	–	2000
[23]	BBN	Levantine dialectal sentences	498	575	126	1199
[24]	SYR (Syria)	Syrian dialectal tweets	448	1350	202	2000
[25]	MD	multidialectal Arabic tweets	734	377	–	1111
[26]	RR	multidialectal Arabic tweets	1752	1941	3698	7391
[Twitter API]	MASTD	multidialectal Arabic tweets	415	777	658	1850
[27]	ArSAS	dialectal sentences (DA)	4643	7840	7279	19,762
[26]	GS	dialectal sentences (DA)	559	1232	2400	4191
[28]	Main-AHS	multidialectal Arabic	628	1398	–	2026
[29]	NU	Egyptian dialectal tweets	1046	976	724	2746
[29]	NBI	Saudi and MSA	2686	3225	3745	9656
[30]	SR	Egyptian dialectal tweets	1612	1604	1604	4820
[31]	LABR	multidialectal Arabic	8224	8224	–	16,448
[32]	MPQA	multidialectal Arabic	4597	5399	–	9996
[33]	Multi-Domain Resource	multidialectal Arabic	24948	6650	–	31,598
[34]	AJGT	Jordanian dialect	720	1080	–	1800
[35]	ASTC	multidialectal Arabic tweets	29849	28902	–	58751