Investigation of bayesian networks for classification problems involving binary data

1 function nb_data_run()

2

3 % Used to do inference testing for NB Data Strcuture

4 % This is the corrected code !!!!!!

5

6 fid=fopen('NB Data Structure Full Summary.txt','w');

7 fprintf(fid,'This is the Full details for NB Data Structure with varying Data Size.');

8 fprintf(fid,'\n -\n') ;

23 fid=fopen('NB Data Structure Full Summary.txt','a');

24 fid1=fopen('NB Data Structure Dataset.txt','a');

25 fid2=fopen('NB Data Structure Summarised Results.csv','a');

82 fprintf(fid,'\nNumber of Nodes %3.0f\n',num_of_nodes);

83 fprintf(fid,'\nRun Number %3.0f\n',run);

84 fprintf(fid,'\n%3.0f training examples\n',train_counter);

85

86 fprintf(fid1,'\nNumber of Nodes %3.0f\n',num_of_nodes);

87 fprintf(fid1,'\nRun Number %3.0f\n',run);

88 fprintf(fid1,'\n%3.0f training examples\n',train_counter);

145 % Calculating Accuracy of Actual Prediction or Bayes Error

146 % *************************************************************************** **************************************************

201 fprintf(fid,'\n%3.0f correct for Actual Structure.\n',correct);

202 fprintf(fid,'\n%3.3f is the Actual Negative Log Likelihood Score.\n',BN_llhoo d);

203 fprintf(fid,'\n%3.0f unknown for Actual Structure.\n',unknown);

210 % Calculating Accuracy of Naive Bayes Prediction

211 % *************************************************************************** **************************************************

262 for row= 1:1:size(xt,1)

263 for col= 1:1:(num_var)

305 fprintf(fid,'\n%3.0f correct for Actual Structure.\n',correct);

306 fprintf(fid,'\n%3.3f is the Actual Negative Log Likelihood Score.\n',BN_llhoo d);

307 fprintf(fid,'\n%3.0f unknown for Actual Structure.\n',unknown);

308 fprintf(fid,'\n%3.0f time for inference and learning for Actual Structure.\n' ,Actual_Time);

309

310 % fprintf(fid1,'\n%3.0f correct for Actual Structure.\n',correct);

311 % fprintf(fid1,'\n%3.3f is the Actual Negative Log Likelihood Score.\n',BN_ll hood);

312 % fprintf(fid1,'\n%3.0f unknown for Actual Structure.\n',unknown);

313 % fprintf(fid1,'\n%3.0f time for inference and learning for Actual Structure \n',Actual_Time);

314

315 fprintf(fid2,'%3.0f,%3.3f,%3.0f,%3.2f,',correct,BN_llhood,unknown,Actual_Time );

320 % *************************************************************************** *******************************************************

321 % *************************************************************************** *******************************************************

334 % Build a Maximumlly Weighted Tree

335 for var1 = 1:1:num_var

336 for var2 = 1:1:num_var

375 % the data has to be fed in with examples in columns

376 % the data to be fed in must be (1,2) and not (1,0)

421 for row= 1:1:size(xt,1)

422 for col= 1:1:(num_var)

466 fprintf(fid,'\n%3.0f correct for TANB Structure.\n',correct);

467 fprintf(fid,'\n%3.3f is the TANB Negative Log Likelihood Score.\n',BN_llhood) ;

468 fprintf(fid,'\n%3.0f unknown for TANB Structure.\n',unknown);

469 fprintf(fid,'\n%3.2f time for learning for TANB Structure.\n',TANB_learn_time );

470 fprintf(fid,'\n%3.2f time for inference for TANB Structure.\n',TANB_infer_tim e);

471 fprintf(fid,'\n%3.2f time for inference and learning for TANB Structure.\n',T ANB_time);

472

473 fprintf(fid2,'%3.0f,%3.3f,%3.0f,%3.2f,%3.2f,%3.2f,',correct,BN_llhood,unknown ,TANB_learn_time,TANB_infer_time,TANB_time);

478 % *************************************************************************** *******************************************************

479 % *************************************************************************** *******************************************************

519 % the data has to be fed in with examples in columns

520 % the data to be fed in must be (1,2) and not (1,0)

567 for row= 1:1:size(xt,1)

568 for col= 1:1:(num_var)

614 fprintf(fid,'\n%3.0f correct for Full BN.\n',correct);

615 fprintf(fid,'\n%3.3f is the BN Negative Log Likelihood Score.\n',BN_llhood);

616 fprintf(fid,'\n%3.0f unknown for FULL BN Structure.\n',unknown);

617 fprintf(fid,'\n%3.2f needed to learn for K2 algo.\n',BN_learn_time);

618 fprintf(fid,'\n%3.2f needed to infer for K2 algo.\n',BN_infer_time);

619 fprintf(fid,'\n%3.2f needed to infer and learn for K2 algo.\n',BN_time);

620 fprintf(fid,'\n************************************************************** *******************\n');

621

622 % fprintf(fid1,'\n%3.0f correct for Full BN.\n',correct);

623 % fprintf(fid1,'\n%3.3f is the BN Negative Log Likelihood Score.\n',BN_llhood );

624 % fprintf(fid1,'\n%3.0f unknown for FULL BN Structure.\n',unknown);

625 % fprintf(fid1,'\n%3.2f needed to learn for K2 algo.\n',BN_learn_time);

626 % fprintf(fid1,'\n%3.2f needed to infer for K2 algo.\n',BN_infer_time);

627 % fprintf(fid1,'\n%3.2f needed to infer and learn for K2 algo.\n',BN_time);

628 % fprintf(fid1,'\n*********************************************************** **********************\n');

629

630 fprintf(fid2,'%3.0f,%3.3f,%3.0f,%3.2f,%3.2f,%3.2f,',correct,BN_llhood,unknown ,BN_learn_time,BN_infer_time,BN_time);

1 function tanb_data_run()

2

3 % Used to do inference testing for TANB Data Strcuture

4 % This is used to test out when ordering is opposite to that stated for BN

5 % This is the corrected code !!!!!!

6

7 fid=fopen('TANB Data Structure Full Summary.txt','w');

8 fprintf(fid,'This is the Full details for TANB Data Structure with varying Data Size.');

9 fprintf(fid,'\n -\n') ;

24 fid=fopen('TANB Data Structure Full Summary.txt','a');

25 fid1=fopen('TANB Data Structure Dataset.txt','a');

26 fid2=fopen('TANB Data Structure Summarised Results.txt','a');

34 dag(2:N-1,1:N-2)=(eye(N-2)); % rot90 rotates the matrix by 90 degrees

35 dag(N,1:(N-1))=1; % Root Node to all nodes

78 fprintf(fid,'\nNumber of Nodes %3.0f\n',num_of_nodes);

79 fprintf(fid,'\nRun Number %3.0f\n',run);

80 fprintf(fid,'\n%3.0f training examples\n',train_counter);

81

82 fprintf(fid1,'\nNumber of Nodes %3.0f\n',num_of_nodes);

83 fprintf(fid1,'\nRun Number %3.0f\n',run);

84 fprintf(fid1,'\n%3.0f training examples\n',train_counter);

145 % Calculating Accuracy of Naive Bayes

146 % *************************************************************************************** **************************************

163 % Creating the Structure of NB

164 % *************************************************************************************** **************************************

188 fprintf(fid,'\nDisplaying Learned Parameters using Actual Network Structure.\n');

210 for row= 1:1:size(xt,1)

211 for col= 1:1:(num_var)

252 fprintf(fid,'\n%3.3f is the Actual Negative Log Likelihood Score.\n',BN_llhood);

253 fprintf(fid,'\n%3.0f unknown for Actual Structure.\n',unknown);

254 fprintf(fid,'\n%3.0f time for inference and learning for Actual Structure.\n',NB_Time);

255

256 % fprintf(fid1,'\n%3.0f correct for Actual Structure.\n',correct);

257 % fprintf(fid1,'\n%3.3f is the Actual Negative Log Likelihood Score.\n',BN_llhood);

258 % fprintf(fid1,'\n%3.0f unknown for Actual Structure.\n',unknown);

259 % fprintf(fid1,'\n%3.0f time for inference and learning for Actual Structure.\n',NB_Time) ;

266 % *************************************************************************************** *******************************************

267 % *************************************************************************************** *******************************************

280 % Build a Maximumlly Weighted Tree

281 for var1 = 1:1:num_var

282 for var2 = 1:1:num_var

321 % the data has to be fed in with examples in columns

322 % the data to be fed in must be (1,2) and not (1,0)

367 for row= 1:1:size(xt,1)

368 for col= 1:1:(num_var)

410 fprintf(fid,'\n%3.0f correct for TANB Structure.\n',correct);

411 fprintf(fid,'\n%3.3f is the TANB Negative Log Likelihood Score.\n',BN_llhood);

412 fprintf(fid,'\n%3.0f unknown for TANB Structure.\n',unknown);

413 fprintf(fid,'\n%3.2f time for learning for TANB Structure.\n',TANB_learn_time);

414 fprintf(fid,'\n%3.2f time for inference for TANB Structure.\n',TANB_infer_time);

415 fprintf(fid,'\n%3.2f time for inference and learning for TANB Structure.\n',TANB_time);

416

417 % fprintf(fid1,'\n%3.0f correct for TANB Structure.\n',correct);

418 % fprintf(fid1,'\n%3.3f is the TANB Negative Log Likelihood Score.\n',BN_llhood);

419 % fprintf(fid1,'\n%3.0f unknown for TANB Structure.\n',unknown);

420 % fprintf(fid1,'\n%3.2f time for learning for TANB Structure.\n',TANB_learn_time);

421 % fprintf(fid1,'\n%3.2f time for inference for TANB Structure.\n',TANB_infer_time);

422 % fprintf(fid1,'\n%3.2f time for inference and learning for TANB Structure.\n',TANB_time) ;

423

424 fprintf(fid2,'%3.0f\t%3.3f\t%3.0f\t%3.2f\t%3.2f\t%3.2f\t',correct,BN_llhood,unknown,TANB_ learn_time,TANB_infer_time,TANB_time);

429 % *************************************************************************************** *******************************************

430 % *************************************************************************************** *******************************************

470 % the data has to be fed in with examples in columns

471 % the data to be fed in must be (1,2) and not (1,0)

518 for row= 1:1:size(xt,1)

519 for col= 1:1:(num_var)

563 fprintf(fid,'\n%3.0f correct for Full BN.\n',correct);

564 fprintf(fid,'\n%3.3f is the BN Negative Log Likelihood Score.\n',BN_llhood);

565 fprintf(fid,'\n%3.0f unknown for FULL BN Structure.\n',unknown);

566 fprintf(fid,'\n%3.2f needed to learn for K2 algo.\n',BN_learn_time);

567 fprintf(fid,'\n%3.2f needed to infer for K2 algo.\n',BN_infer_time);

568 fprintf(fid,'\n%3.2f needed to infer and learn for K2 algo.\n',BN_time);

569 fprintf(fid,'\n************************************************************************** *******\n');

570

571 % fprintf(fid1,'\n%3.0f correct for Full BN.\n',correct);

572 % fprintf(fid1,'\n%3.3f is the BN Negative Log Likelihood Score.\n',BN_llhood);

573 % fprintf(fid1,'\n%3.0f unknown for FULL BN Structure.\n',unknown);

574 % fprintf(fid1,'\n%3.2f needed to learn for K2 algo.\n',BN_learn_time);

575 % fprintf(fid1,'\n%3.2f needed to infer for K2 algo.\n',BN_infer_time);

576 % fprintf(fid1,'\n%3.2f needed to infer and learn for K2 algo.\n',BN_time);

577 % fprintf(fid1,'\n*********************************************************************** **********\n');

578

579 fprintf(fid2,'%3.0f\t%3.3f\t%3.0f\t%3.2f\t%3.2f\t%3.2f\t',correct,BN_llhood,unknown,BN_le arn_time,BN_infer_time,BN_time);

1 function bn_data_run_20()

2

3 % Used to do inference testing for BN Data Strcuture for 100 Nodes

4 % This code involves noise in both training and testing sets

5 % This is the corrected code !!!!!!

6

7 fid=fopen('BN Data Structure Full Summary.txt','w');

8 fprintf(fid,'This is the Full details for NB Data Structure with varying Data Size.');

9 fprintf(fid,'\n -\n') ;

25 fid=fopen('BN Data Structure Full Summary.txt','a');

26 fid1=fopen('BN Data Structure Dataset.txt','a');

27 fid2=fopen('BN Data Structure Summarised Results.csv','a');

90 fprintf(fid,'\nNumber of Nodes %3.0f\n',num_of_nodes);

91 fprintf(fid,'\nRun Number %3.0f\n',run);

92 fprintf(fid,'\n%3.0f training examples\n',train_counter);

93

94 fprintf(fid1,'\nNumber of Nodes %3.0f\n',num_of_nodes);

95 fprintf(fid1,'\nRun Number %3.0f\n',run);

96 fprintf(fid1,'\n%3.0f training examples\n',train_counter);

126 train_filename = strcat(nodes_string,'_nodes_',num_of_links_string,'_max_ links_',run_string,'_run_',train_string,'_train_num_TRAIN.csv');

148 dlmwrite(test_filename,initial_gen_tdata');

149

150 % % ********************************************************************* ********************************************************

151 % % Calculating Accuracy of Actual Prediction or Bayes Error

152 % % ********************************************************************* ********************************************************

207 % fprintf(fid,'\n%3.0f correct for Actual Structure.\n',correct);

208 % fprintf(fid,'\n%3.3f is the Actual Negative Log Likelihood Score.\n',BN _llhood);

209 % fprintf(fid,'\n%3.0f unknown for Actual Structure.\n',unknown);

216 % Calculating Accuracy of Naive Bayes Prediction

217 % *********************************************************************** ******************************************************

270 for row= 1:1:size(xt,1)

271 for col= 1:1:(num_var)

313 fprintf(fid,'\n%3.0f correct for Actual Structure.\n',correct);

314 fprintf(fid,'\n%3.3f is the Actual Negative Log Likelihood Score.\n',BN_l lhood);

315 fprintf(fid,'\n%3.0f unknown for Actual Structure.\n',unknown);

316 fprintf(fid,'\n%3.0f time for inference and learning for Actual Structure .\n',Actual_Time);

317

318 % fprintf(fid1,'\n%3.0f correct for Actual Structure.\n',correct);

319 % fprintf(fid1,'\n%3.3f is the Actual Negative Log Likelihood Score.\n',B N_llhood);

320 % fprintf(fid1,'\n%3.0f unknown for Actual Structure.\n',unknown);

321 % fprintf(fid1,'\n%3.0f time for inference and learning for Actual Struct ure.\n',Actual_Time);

322

323 fprintf(fid2,'%3.0f,%3.3f,%3.0f,%3.2f,',correct,BN_llhood,unknown,Actual_ Time);

328 % *********************************************************************** ***********************************************************

329 % *********************************************************************** ***********************************************************

342 % Build a Maximumlly Weighted Tree

343 for var1 = 1:1:num_var

344 for var2 = 1:1:num_var

383 % the data has to be fed in with examples in columns

384 % the data to be fed in must be (1,2) and not (1,0)

429 for row= 1:1:size(xt,1)

430 for col= 1:1:(num_var)

474 fprintf(fid,'\n%3.0f correct for TANB Structure.\n',correct);

475 fprintf(fid,'\n%3.3f is the TANB Negative Log Likelihood Score.\n',BN_llh ood);

476 fprintf(fid,'\n%3.0f unknown for TANB Structure.\n',unknown);

477 fprintf(fid,'\n%3.2f time for learning for TANB Structure.\n',TANB_learn_ time);

478 fprintf(fid,'\n%3.2f time for inference for TANB Structure.\n',TANB_infer _time);

479 fprintf(fid,'\n%3.2f time for inference and learning for TANB Structure.\ n',TANB_time);

480

481 fprintf(fid2,'%3.0f,%3.3f,%3.0f,%3.2f,%3.2f,%3.2f,',correct,BN_llhood,unk nown,TANB_learn_time,TANB_infer_time,TANB_time);

482

483 clear x y xt yt engine evidence bnet1 correct unknown BN_llhood TANB_lear n_time TANB_infer_time TANB_time;

484

485 % *********************************************************************** ***********************************************************

486 % *********************************************************************** ***********************************************************

487 % *********************************************************************** ***********************************************************

527 % the data has to be fed in with examples in columns

528 % the data to be fed in must be (1,2) and not (1,0)

575 for row= 1:1:size(xt,1)

576 for col= 1:1:(num_var)

622 fprintf(fid,'\n%3.0f correct for Full BN.\n',correct);

623 fprintf(fid,'\n%3.3f is the BN Negative Log Likelihood Score.\n',BN_llhoo d);

624 fprintf(fid,'\n%3.0f unknown for FULL BN Structure.\n',unknown);

625 fprintf(fid,'\n%3.2f needed to learn for K2 algo.\n',BN_learn_time);

626 fprintf(fid,'\n%3.2f needed to infer for K2 algo.\n',BN_infer_time);

627 fprintf(fid,'\n%3.2f needed to infer and learn for K2 algo.\n',BN_time);

628 fprintf(fid,'\n********************************************************** ***********************\n');

629

630 % fprintf(fid1,'\n%3.0f correct for Full BN.\n',correct);

631 % fprintf(fid1,'\n%3.3f is the BN Negative Log Likelihood Score.\n',BN_ll hood);

632 % fprintf(fid1,'\n%3.0f unknown for FULL BN Structure.\n',unknown);

633 % fprintf(fid1,'\n%3.2f needed to learn for K2 algo.\n',BN_learn_time);

634 % fprintf(fid1,'\n%3.2f needed to infer for K2 algo.\n',BN_infer_time);

635 % fprintf(fid1,'\n%3.2f needed to infer and learn for K2 algo.\n',BN_time );

636 % fprintf(fid1,'\n******************************************************* **************************\n');

637

638 fprintf(fid2,'%3.0f,%3.3f,%3.0f,%3.2f,%3.2f,%3.2f,',correct,BN_llhood,unk nown,BN_learn_time,BN_infer_time,BN_time);

1 function nb_data_run_noise()

2

3 % Used to do inference testing for NB Data Strcuture

4 % This code involves noise in both training and testing sets

5 % This is the corrected code !!!!!!

6

7 fid=fopen('NB Data Structure Full Summary.txt','w');

8 fprintf(fid,'This is the Full details for NB Data Structure with varying Data Size.');

9 fprintf(fid,'\n -\n') ;

24 fid=fopen('NB Data Structure Full Summary.txt','a');

25 fid1=fopen('NB Data Structure Dataset.txt','a');

26 fid2=fopen('NB Data Structure Summarised Results.csv','a');

63 % To Display the Actual Parameters

83 fprintf(fid,'\nNumber of Nodes %3.0f\n',num_of_nodes);

84 fprintf(fid,'\nRun Number %3.0f\n',run);

85 fprintf(fid,'\n%3.0f training examples\n',train_counter);

86

87 fprintf(fid1,'\nNumber of Nodes %3.0f\n',num_of_nodes);

88 fprintf(fid1,'\nRun Number %3.0f\n',run);

89 fprintf(fid1,'\n%3.0f training examples\n',train_counter);

149 dlmwrite(test_filename,initial_gen_tdata');

150

151 % *************************************************************************** **************************************************

152 % Calculating Accuracy of Actual Prediction or Bayes Error

153 % *************************************************************************** **************************************************

208 fprintf(fid,'\n%3.0f correct for Actual Structure.\n',correct);

209 fprintf(fid,'\n%3.3f is the Actual Negative Log Likelihood Score.\n',BN_llhoo d);

210 fprintf(fid,'\n%3.0f unknown for Actual Structure.\n',unknown);

217 % Calculating Accuracy of Naive Bayes Prediction

218 % *************************************************************************** **************************************************

269 for row= 1:1:size(xt,1)

270 for col= 1:1:(num_var)

311

312 fprintf(fid,'\n%3.0f correct for Actual Structure.\n',correct);

313 fprintf(fid,'\n%3.3f is the Actual Negative Log Likelihood Score.\n',BN_llhoo d);

314 fprintf(fid,'\n%3.0f unknown for Actual Structure.\n',unknown);

315 fprintf(fid,'\n%3.0f time for inference and learning for Actual Structure.\n' ,Actual_Time);

316

317 % fprintf(fid1,'\n%3.0f correct for Actual Structure.\n',correct);

318 % fprintf(fid1,'\n%3.3f is the Actual Negative Log Likelihood Score.\n',BN_ll hood);

319 % fprintf(fid1,'\n%3.0f unknown for Actual Structure.\n',unknown);

320 % fprintf(fid1,'\n%3.0f time for inference and learning for Actual Structure \n',Actual_Time);

321

322 fprintf(fid2,'%3.0f,%3.3f,%3.0f,%3.2f,',correct,BN_llhood,unknown,Actual_Time );

327 % *************************************************************************** *******************************************************

328 % *************************************************************************** *******************************************************

341 % Build a Maximumlly Weighted Tree

342 for var1 = 1:1:num_var

343 for var2 = 1:1:num_var

382 % the data has to be fed in with examples in columns

383 % the data to be fed in must be (1,2) and not (1,0)

428 for row= 1:1:size(xt,1)

429 for col= 1:1:(num_var)