WEBVTT 00:01.050 --> 00:07.800 All right, anyway, so let's try to grab plus 30 percentage of sentences, which is having a maximum 00:07.800 --> 00:11.550 score out of this sentence code dictionary we helped create. 00:12.480 --> 00:17.640 So for that, we are going to use this from Tipu. 00:20.760 --> 00:22.980 Import and largest. 00:25.980 --> 00:32.640 Let's first grab how many total sentences that that sentence caught hiding lengthens function will work 00:32.640 --> 00:33.630 on a dictionary or so. 00:34.180 --> 00:35.690 So that will be a 24th. 00:35.910 --> 00:41.130 And if we just grab zero point three of that, that will be a. 00:42.290 --> 00:43.410 So point nineteen. 00:43.550 --> 00:46.250 That means maximum eight sentences. 00:46.340 --> 00:47.630 We can grab it. 00:47.990 --> 00:50.870 So let's try to grab first a sentence. 00:51.140 --> 00:53.480 So let's use and the largest. 00:54.950 --> 00:57.530 So the first argument, we have to pass it. 00:58.010 --> 01:00.620 And that how many of them you want to. 01:01.890 --> 01:02.410 Selective. 01:02.780 --> 01:08.530 Let's just go with eight, which is terrible, on which you are going to do all those things. 01:09.500 --> 01:11.990 So any terrible will be sentenced. 01:12.090 --> 01:12.390 Court. 01:15.660 --> 01:16.930 Let me make it here. 01:17.280 --> 01:23.550 And it's great and based on which key you are going to do all those things. 01:24.070 --> 01:26.000 So he will be sentenced. 01:28.310 --> 01:29.700 He thought you get. 01:30.690 --> 01:38.100 Now, here you are not calling any function instead of the get function will return as those value. 01:38.460 --> 01:41.610 And let me assign it to somebody. 01:42.000 --> 01:44.040 Let me bring somebody. 01:47.280 --> 01:48.270 So these are the. 01:49.410 --> 01:51.670 Summarization of our own article. 01:52.050 --> 02:00.390 So first, let's try to combine all those sentences together and we will put it into one single bucket. 02:00.780 --> 02:04.920 So far they were not tax. 02:06.500 --> 02:10.230 Far more in somebody's. 02:12.200 --> 02:18.020 And that will be a sign and underscore somebody. 02:20.730 --> 02:21.880 Let me bring Dick. 02:29.290 --> 02:30.610 And that will be a part of. 02:32.610 --> 02:41.640 Now we can even simple use like a joint space on this final somebody. 02:44.230 --> 02:45.780 And let me just misplay. 02:48.460 --> 02:51.220 So that is our complete final somebody. 02:51.760 --> 02:54.650 Let me make it into radio. 02:54.820 --> 02:55.960 Let's hear somebody tell. 02:59.090 --> 02:59.810 Let me bring 03:03.380 --> 03:04.340 somebody video, but. 03:06.770 --> 03:07.180 All right. 03:07.360 --> 03:09.850 So that is the summarization of the whole article. 03:10.270 --> 03:16.450 Now we can do anything, we can just simply find a land of this somebody. 03:18.980 --> 03:19.910 That will be a six. 03:20.030 --> 03:22.010 Thirty two characters. 03:22.340 --> 03:27.280 And if you just try to divide this thing by our holy text. 03:28.610 --> 03:29.850 So that will be a. 03:35.680 --> 03:38.400 It's almost 40 percent is your data. 03:38.960 --> 03:41.230 We gather as a summarization. 03:41.870 --> 03:45.350 Almost 60 percent range of data to reduce. 03:45.500 --> 03:45.860 Really? 03:46.550 --> 03:49.130 So that is what the summary of our whole article. 03:49.550 --> 03:52.070 But this is just a very simple mechanism. 03:52.160 --> 03:55.510 We have applied for automated text summarization. 03:55.850 --> 03:56.490 All right. 03:56.600 --> 03:59.230 So that is the whole idea behind the text summarization. 03:59.540 --> 04:00.650 See you in the next video. 04:00.650 --> 04:04.220 Will see some more stuff related to natural language processing.