NGHIÊN CỨU XÂY DỰNG HỆ THỐNG TRẢ LỜI TỰ ĐỘNG CHUYÊN NGÀNH CÔNG NGHỆ THÔNG TIN - Full 10 điểm

15 0 0
NGHIÊN CỨU XÂY DỰNG HỆ THỐNG TRẢ LỜI TỰ ĐỘNG CHUYÊN NGÀNH CÔNG NGHỆ THÔNG TIN - Full 10 điểm

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

Thông tin tài liệu

TRƯ Ờ NG Đ Ạ I H Ọ C NAM C Ầ N THƠ KHOA K Ỹ THU Ậ T – CÔNG NGH Ệ TRƯƠNG MINH C Ả NH 1 76892 NGHIÊN C Ứ U XÂY D Ự NG H Ệ TH Ố NG TR Ả L Ờ I T Ự Đ Ộ NG CHUYÊN NGÀNH CÔNG NGH Ệ THÔNG TIN Đ Ồ ÁN TH Ự C T Ậ P Ngành Công ngh ệ Thông tin Mã s ố Ngành: 7480201 C ầ n Thơ, tháng 05 năm 2021 TRƯ Ờ NG Đ Ạ I H Ọ C NAM C Ầ N THƠ KHOA K Ỹ THU Ậ T – CÔNG NGH Ệ TRƯƠNG MINH C Ả NH MSSV: 176 8 92 NGHIÊN C Ứ U XÂY D Ự NG H Ệ TH Ố NG TR Ả L Ờ I T Ự Đ Ộ NG CHUYÊN NGÀNH CÔNG NGH Ệ THÔNG TIN Đ Ồ ÁN TH Ự C T Ậ P Ngành Công ngh ệ Thông tin Mã s ố Ngành: 7480201 GI Ả NG VIÊN HƯ Ớ NG D Ẫ N TS NGÔ H Ồ ANH KHÔI C ầ n Th ơ, th áng 05 n ăm 2021 i NH Ậ N XÉT C Ủ A C Ơ QUAN TH Ự C T Ậ P ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· C ầ n Thơ, Ngày… tháng… năm 2021 C án b ộ hư ớ ng d ẫ n (Ký tên) PH Ạ M TH Ị TH Ủ Y ii NH Ậ N XÉT C Ủ A GIÁO VIÊN HƯ Ớ NG D Ẫ N ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· ································ ································ ································ ···· C ầ n Thơ, Ngày… tháng… năm 2021 Giáo viên hư ớ ng d ẫ n (Ký tên) T S Ngô H ồ Anh Khôi iii NH Ậ N XÉT C Ủ A GIÁO VIÊN PH Ả N BI Ệ N ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· ································ ································ ································ ····· C ầ n Thơ, Ngày… tháng… năm 2021 Giáo viên ph ả n bi ệ n (Ký tên) ThS Hu ỳ nh Bá l ộ c iv L Ờ I C Ả M ƠN Tôi xin chân thành g ử i l ờ i c ả m ơn đ ế n TS Ngô H ồ Anh Khôi - Gi ả ng viên Trư ờ ng Đ ạ i h ọ c Nam C ầ n Thơ đã nhi ệ t tình hư ớ ng d ẫ n trong su ố t th ờ i gian làm đ ề tài th ự c t ậ p t ố t nghi ệ p Th ầ y là ngư ờ i đã t ạ o ra r ấ t nhi ề u đi ề u ki ệ n thu ậ n l ợ i và cho em nh ữ ng l ờ i khuyên quý báu nh ằ m giúp chúng em hoàn thành t ố t đ ề tài th ự c t ậ p t ố t nghi ệ p này Xin g ử i l ờ i c ả m ơn chân thành đ ế n th ầ y cô giáo trong Trư ờ ng Đ ạ i h ọ c Nam C ầ n Thơ nói chung và các th ầ y cô trong B ộ môn Công ngh ệ thông tin nói riêng đã nhi ệ t tình gi ả ng d ạ y, truy ề n đ ạ t nh ữ ng ki ế n th ứ c uyên thâm và t ạ o cho em nhi ề u đi ề u ki ệ n đ ể giúp đ ỡ trong su ố t quá trình theo h ọ c t ạ i trư ờ ng nh ữ ng năm h ọ c qua, giúp em có đư ợ c nh ữ ng cơ s ở lý thuy ế t v ữ ng vàng đ ể có th ể th ự c hi ệ n đ ề tài ngày hôm nay L ờ i cu ố i, em xin chân thành tha thi ế t g ử i l ờ i c ả m ơn đ ế n b ạ n bè và gia đình, đã luôn quan tâm, t ạ o đi ề u ki ệ n khi em c ầ n, giúp đ ỡ , ủ ng h ộ và đ ộ ng viên em trong quá trình h ọ c t ậ p và hoàn thành đ ồ án th ự c t ậ p t ố t nghi ệ p Xin chân t hành c ả m ơn! C ầ n Thơ, ngày … tháng … năm 2021 Sinh viên th ự c hi ệ n v L Ờ I CAM ĐOAN Tôi xin cam k ế t r ằ ng đây là m ộ t s ả n ph ẩ m hoàn toàn do quá trình h ọ c t ậ p và v ậ n d ụ ng ki ế n th ứ c đã đư ợ c h ọ c t ạ i Trư ờ ng Đ ạ i h ọ c Nam C ầ n Thơ đ ể hoàn thành s ả n ph ẩ m c ủ a riêng c ủ a mình trong khuôn kh ổ c ủ a đ ề tài V ề hình th ứ c, nh ữ ng k ế t lu ậ n đư ợ c trình bày trong đ ồ án th ự c t ậ p này là hoàn toàn trung th ự c C ầ n Thơ, ngày … tháng … năm 2021 Sinh viên th ự c hi ệ n vi M Ụ C L Ụ C NH Ậ N XÉT C Ủ A CƠ QUAN TH Ự C T Ậ P i NH Ậ N XÉT C Ủ A GIÁO VIÊN HƯ Ớ NG D Ẫ N ii NH Ậ N XÉT C Ủ A GIÁO VIÊN PH Ả N BI Ệ N iii L Ờ I C Ả M ƠN iv L Ờ I CAM ĐOAN v M Ụ C L Ụ C vi TÓM T Ắ T xiv CHƯƠNG 1 GI Ớ I THI Ệ U 1 1 1 T ổ ng quan v ề cơ quan th ự c t ậ p 1 1 1 1 Thông tin v ề công ty 1 1 1 2 Tr ụ s ở chính 1 1 1 3 Ngư ờ i đ ạ i di ệ n theo pháp lu ậ t 1 1 1 4 Thông tin li ệ n h ệ 1 1 1 5 Ngành ngh ề kinh doanh 2 1 1 6 S ả n ph ẩ m và d ị ch v ụ đang kinh doanh 3 1 1 6 1 S ả n ph ẩ m ph ầ n m ề m 3 1 1 6 2 D ị ch v ụ 7 1 2 Gi ớ i thi ệ u chung 8 1 2 1 Đ ặ t v ấ n đ ề nghiên c ứ u 8 1 2 2 M ụ c tiêu nghiên c ứ u 8 1 2 2 1 M ụ c tiêu chung 8 1 2 2 2 M ụ c tiêu c ụ th ể 9 1 2 3 Ph ạ m vi nghiên c ứ u 9 1 2 3 1 Không gian 9 1 2 3 2 Th ờ i gian 9 1 2 3 3 Đ ố i tư ợ ng nghiên c ứ u 9 1 2 4 Ph ạ m vi đ ề tài 9 1 2 5 Hư ớ ng gi ả i quy ế t v ấ n đ ề 9 1 2 6 B ố c ụ c 10 vii CHƯƠNG 2 CƠ S Ở LÝ THUY Ế T VÀ PHƯƠNG PHÁP NGHIÊN C Ứ U 11 2 1 Cơ s ở lý thuy ế t 11 2 2 1 Trí tu ệ nhân t ạ o 11 2 1 1 1 Khái ni ệ m 11 2 1 1 2 Phân lo ạ i 12 2 1 1 3 Ứ ng d ụ ng 13 2 1 2 Chatbots 16 2 1 2 1 Khái ni ệ m 16 2 1 2 2 L ị ch s ử hình thành 16 2 1 2 3 C ấ u t ạ o chatbots 21 2 1 2 4 Phân lo ạ i 22 2 1 2 5 Cách th ứ c ho ạ t đ ộ ng 23 2 1 2 6 M ộ t s ố n ề n t ả ng h ỗ tr ợ cho vi ệ c phát tri ể n chatbots 23 2 1 2 7 M ộ t s ố ứ ng d ụ ng chatbot 24 2 1 2 8 Xu hư ớ ng phát tri ể n 24 2 1 3 X ử lý ngôn ng ữ t ự nhiên 25 2 1 3 1 Khái ni ệ m 25 2 1 3 2 Các bư ớ c x ử lý 25 2 1 3 3 Các bài toán ứ ng d ụ ng 25 2 1 3 4 Các bài toán x ử lý 27 2 2 Phương pháp nghiên c ứ u 27 2 2 1 Phương pháp nghiên c ứ u lý thuy ế t 27 2 2 2 Phương pháp nghiên c ứ u th ự c nghi ệ m 28 2 2 3 Phương pháp đi ề u tra 28 2 3 Gi ớ i thi ệ u v ề gi ả i Conditional Random Field 28 2 3 1 G i ớ i thi ệ u chung 28 2 3 2 Mô t ả 29 2 3 3 Bi ế n th ể 29 2 4 Gi ớ i thi ệ u v ề ngôn ng ữ Python 30 viii 2 4 1 Khái ni ệ m 30 2 4 2 L ị ch s ử 30 2 4 3 Tính năng 31 2 4 4 Đư ợ c ứ ng d ụ ng 31 CHƯƠNG 3 GI Ả I THU Ậ T CRF VÀ CÁC CH I Ế N LƯ Ợ C ÁP D Ụ NG 33 3 1 Trích xu ấ t tính năng văn b ả n và tài li ệ u 33 3 1 1 Mã hóa 33 3 1 2 Ngưng t ừ 33 3 1 3 Vi ế t hoa 33 3 1 4 Ti ế ng lóng và t ừ vi ế t t ắ c 33 3 1 5 Lo ạ i b ỏ ti ế ng ồ n 34 3 1 6 S ử a chính t ả 34 3 1 7 G ố c 34 3 1 8 B ổ sung 34 3 1 9 Word2Vec 35 3 1 10 Vectơ toàn c ầ u cho bi ể u di ễ n t ừ (GloVe) 35 3 1 11 Trình bày t ừ ng ữ theo ng ữ c ả nh 35 3 1 12 FastText 37 3 1 13 T ầ n su ấ t k ỳ h ạ n 37 3 1 14 T ầ n su ấ t tài li ệ u ngh ị ch đ ả o thu ậ t ng ữ 37 3 2 Gi ả m kích thư ớ c 38 3 3 K ỹ thu ậ t phân lo ạ i văn b ả n 38 3 3 1 Thúc đ ẩ y 39 3 3 2 Đóng bao 39 3 3 3 M ạ ng th ầ n kinh sâu 41 3 3 4 M ạ ng th ầ n kinh tái di ễ n (RNN) 41 3 3 5 M ạ ng th ầ n kinh h ợ p pháp (CNN) 42 3 3 6 M ạ ng th ầ n kinh chuy ể n đ ổ i l ặ p l ạ i (RCNN) 42 3 3 7 H ọ c s âu đa mô hình ng ẫ u nhiên (RMDL) 42 ix 3 3 8 H ọ c sâu phân c ấ p cho văn b ả n (HDLTex) 43 3 4 Đánh giá 43 3 5 Văn b ả n và b ộ d ữ li ệ u tài li ệ u 44 3 6 Ứ ng d ụ ng phân lo ạ i văn b ả n 45 3 7 H ỗ tr ợ phân lo ạ i văn b ả n 46 CHƯƠNG 4 MÔ HÌNH H Ệ TH Ố NG 48 4 1 T ổ ng quan 48 4 1 1 Khái quát 48 4 1 2 Sơ lư ợ c l ị ch s ử phát tri ể n 48 4 1 3 Ki ế n trúc chung c ủ a h ệ th ố ng h ỏ i đáp 48 4 1 3 1 Giao di ệ n ngư ờ i dùng (User Interface) 49 4 1 3 2 Ph ân tích câu h ỏ i (Question Analyzer) 49 4 1 3 3 Tìm ki ế m d ữ li ệ u (Data Retrieval) 49 4 1 3 4 Rút trích câu tr ả l ờ i (Answer Extraction) 49 4 1 3 5 Chi ế n lư ợ c x ế p h ạ ng (Ranking) 50 4 1 3 6 Xác minh câu tr ả l ờ i (Answer Verification) 50 4 1 4 M ộ t s ố v ấ n đ ề quan tâm khi thi ế t k ế h ệ th ố ng h ỏ i đáp 50 4 1 5 Các yêu c ầ u c ủ a h ệ th ố ng h ỏ i đáp 50 4 2 Ti ề n x ử lý và phân lo ạ i câu h ỏ i 50 4 2 1 Ti ề n x ử lý d ữ li ệ u 50 4 2 1 1 Phân tích câu h ỏ i 50 4 2 1 2 Tách t ừ 51 4 2 1 3 Bi ể u di ễ n văn b ả n 52 4 2 2 Phân lo ạ i câu h ỏ i 52 4 2 2 1 Rút trích đ ặ c trưng văn b ả n 52 4 2 2 2 Phân lo ạ i câu h ỏ i ti ế ng Vi ệ t 54 4 3 Xây d ự ng mô hình h ệ th ố ng 55 4 3 1 L ự a ch ọ n mô hình và thu ậ t toán 55 4 3 2 Mô hình ki ế n trúc h ệ th ố ng 56 x 4 3 3 Cài đ ặ t 58 CHƯƠNG 5 K Ế T QU Ả NGHIÊN C Ứ U 59 5 1 H ệ th ố ng 59 5 1 1 Thao tác v ớ i b ộ d ữ li ệ u chuyên ngành Công Ngh ệ Thông Tin 59 5 1 2 Giao di ệ n h ệ th ố ng chương trình 60 5 2 Hư ớ ng d ẫ n cài đ ặ t 62 5 3 Hư ớ ng d ẫ n s ử d ụ ng 63 5 3 1 Thêm d ữ li ệ u 63 5 3 2 Th ự c thi chương trình 64 CHƯƠNG 6 K Ế T LU Ậ N 67 TÀI LI Ệ U THAM KH Ả O 68 xi DANH S ÁCH B Ả NG B ả ng 1 1 Ngành ngh ề kinh doanh 2 B ả ng 1 2 B ả ng s ả n ph ẩ m ph ầ n m ề m 3 B ả ng 1 3 B ả ng chi ti ế t d ị ch v ụ 7 xii DANH SÁCH HÌNH Hình 2 1 Trí thông minh nhân t ạ o 11 Hình 2 2 Mô hình xe không ngư ờ i lái 12 Hình 2 3 Công ngh ệ nh ậ n th ứ c 13 Hình 2 4 Các hình th ứ c giao hàng c ủ a Uber 14 Hình 2 5 Máy bay thi ế t b ị không ngư ờ i láy 14 Hình 2 6 Gi ả i pháp trong giáo d ụ c 15 Hình 2 7 Cách th ứ c ho ạ t đ ộ ng c ủ a Chatbots 16 Hình 2 8 Phép th ử Turing 17 Hình 2 9 Tương tác gi ữ a ngư ờ i dùng v ớ i ELIZA 18 Hình 2 10 Mô hình cách th ứ c ho ạ t đ ộ ng c ủ a Chatbot 23 Hình 4 1 H ệ th ố ng tìm ki ế m thông tin IR 48 Hình 4 2 Ki ế n trúc chung h ệ th ố ng h ỏ i đáp 4 9 Hình 4 3 Mô hình ho ạ t đ ộ ng h ệ th ố ng 56 Hình 4 4 Giai đo ạ n phân tích câu h ỏ i 56 Hình 4 5 Ti ế n trình phân l ớ p 57 Hình 4 6 Ti ế n trình tìm ki ế m câu h ỏ i 57 Hình 4 7 Ti ế n trình tìm ki ế m câu tr ả l ờ i 57 Hình 4 8 Ti ế n trình tìm ki ế m câu h ỏ i đ ế n chuyên gia 57 Hình 5 1 Giao di ệ n chính c ủ a chương trình 60 Hình 5 2 Giao di ệ n chính dành cho qu ả n tr ị viên 60 Hình 5 3 Giao di ệ n thêm d ữ li ệ u 61 Hình 5 4 Giao di ệ n ch ọ n t ệ p d ữ li ệ u c ầ n thêm vào h ệ th ố ng 61 Hình 5 5 Giao di ệ n ch ọ n thư m ụ c ch ứ a các t ệ p c ầ n thêm vào h ệ th ố ng 62 Hình 5 6 T ệ p TraLoiTuDOngCNTT exe 62 Hình 5 7 C ử a s ổ m ớ i cài đ ặ t 62 Hình 5 8 Bi ể u tư ợ ng ch ạ y ch ương trình ở Desktop 63 Hình 5 9 T ệ p ch ạ y chương trình trong thư m ụ c đã cài đ ặ t 63 Hình 5 10 Menu “Thêm d ữ li ệ u” 63 Hình 5 11 Giao di ệ n ch ọ n đư ờ ng d ẫ n c ủ a t ệ p d ữ li ệ u c ầ n thêm vào 64 Hình 5 12 Giao di ệ n chính c ủ a form ngư ờ i dùng 65 Hình 5 13 Giao di ệ n sau khi th ự c thi 65 Hình 5 14 Giao di ệ n c ủ a nhân viên khi ngư ờ i dùng g ử i yêu c ầ u 66 Hình 5 15 Giao di ệ n ngư ờ i dùng sao khi nhân viên ph ả n h ồ i 66 xiii DANH M Ụ C T Ừ VI Ế T T Ắ T T ê n vi ế t t ắ t T ên đ ầ y đ ủ TNHH Trách nhi ệ m h ữ u h ạ n CSDL Cơ s ở d ữ li ệ u NLP Natural Language Processing HOG Histogram of Oriented Gradients OCR Optical Character Recognition CRFs Conditional Random Field AI Artificial intelligence

TRƯỜNG ĐẠI HỌC NAM CẦN THƠ KHOA KỸ THUẬT – CÔNG NGHỆ TRƯƠNG MINH CẢNH 176892 NGHIÊN CỨU XÂY DỰNG HỆ THỐNG TRẢ LỜI TỰ ĐỘNG CHUYÊN NGÀNH CÔNG NGHỆ THÔNG TIN ĐỒ ÁN THỰC TẬP Ngành Công nghệ Thông tin Mã số Ngành: 7480201 Cần Thơ, tháng 05 năm 2021 TRƯỜNG ĐẠI HỌC NAM CẦN THƠ KHOA KỸ THUẬT – CÔNG NGHỆ TRƯƠNG MINH CẢNH MSSV: 176892 NGHIÊN CỨU XÂY DỰNG HỆ THỐNG TRẢ LỜI TỰ ĐỘNG CHUYÊN NGÀNH CÔNG NGHỆ THÔNG TIN ĐỒ ÁN THỰC TẬP Ngành Công nghệ Thông tin Mã số Ngành: 7480201 GIẢNG VIÊN HƯỚNG DẪN TS NGÔ HỒ ANH KHÔI Cần Thơ, tháng 05 năm 2021 NHẬN XÉT CỦA CƠ QUAN THỰC TẬP ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· Cần Thơ, Ngày… tháng… năm 2021 Cán hướng dẫn (Ký tên) PHẠM THỊ THỦY i NHẬN XÉT CỦA GIÁO VIÊN HƯỚNG DẪN ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· ···································································································· Cần Thơ, Ngày… tháng… năm 2021 Giáo viên hướng dẫn (Ký tên) TS Ngô Hồ Anh Khôi ii NHẬN XÉT CỦA GIÁO VIÊN PHẢN BIỆN ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· ····································································································· Cần Thơ, Ngày….tháng… năm 2021 Giáo viên phản biện (Ký tên) ThS Huỳnh Bá lộc iii LỜI CẢM ƠN Tôi xin chân thành gửi lời cảm ơn đến TS Ngô Hồ Anh Khôi - Giảng viên Trường Đại học Nam Cần Thơ nhiệt tình hướng dẫn suốt thời gian làm đề tài thực tập tốt nghiệp Thầy người tạo nhiều điều kiện thuận lợi cho em lời khuyên quý báu nhằm giúp chúng em hoàn thành tốt đề tài thực tập tốt nghiệp Xin gửi lời cảm ơn chân thành đến thầy cô giáo Trường Đại học Nam Cần Thơ nói chung thầy Bộ mơn Cơng nghệ thơng tin nói riêng nhiệt tình giảng dạy, truyền đạt kiến thức uyên thâm tạo cho em nhiều điều kiện để giúp đỡ suốt trình theo học trường năm học qua, giúp em có sở lý thuyết vững vàng để thực đề tài ngày hôm Lời cuối, em xin chân thành tha thiết gửi lời cảm ơn đến bạn bè gia đình, ln quan tâm, tạo điều kiện em cần, giúp đỡ, ủng hộ động viên em trình học tập hoàn thành đồ án thực tập tốt nghiệp Xin chân thành cảm ơn! Cần Thơ, ngày … tháng … năm 2021 Sinh viên thực iv LỜI CAM ĐOAN Tôi xin cam kết sản phẩm hồn tồn q trình học tập vận dụng kiến thức học Trường Đại học Nam Cần Thơ để hoàn thành sản phẩm riêng khn khổ đề tài Về hình thức, kết luận trình bày đồ án thực tập hoàn toàn trung thực Cần Thơ, ngày … tháng … năm 2021 Sinh viên thực v MỤC LỤC NHẬN XÉT CỦA CƠ QUAN THỰC TẬP i NHẬN XÉT CỦA GIÁO VIÊN HƯỚNG DẪN ii NHẬN XÉT CỦA GIÁO VIÊN PHẢN BIỆN iii LỜI CẢM ƠN iv LỜI CAM ĐOAN .v MỤC LỤC vi TÓM TẮT xiv CHƯƠNG GIỚI THIỆU 1.1 Tổng quan quan thực tập 1.1.1 Thông tin công ty 1.1.2 Trụ sở 1.1.3 Người đại diện theo pháp luật 1.1.4 Thông tin liện hệ 1.1.5 Ngành nghề kinh doanh .2 1.1.6 Sản phẩm dịch vụ kinh doanh 1.1.6.1 Sản phẩm phần mềm 1.1.6.2 Dịch vụ 1.2 Giới thiệu chung 1.2.1 Đặt vấn đề nghiên cứu 1.2.2 Mục tiêu nghiên cứu 1.2.2.1 Mục tiêu chung .8 1.2.2.2 Mục tiêu cụ thể .9 1.2.3 Phạm vi nghiên cứu 1.2.3.1 Không gian .9 1.2.3.2 Thời gian 1.2.3.3 Đối tượng nghiên cứu 1.2.4 Phạm vi đề tài 1.2.5 Hướng giải vấn đề 1.2.6 Bố cục 10 vi CHƯƠNG CƠ SỞ LÝ THUYẾT VÀ PHƯƠNG PHÁP NGHIÊN CỨU 11 2.1 Cơ sở lý thuyết 11 2.2.1 Trí tuệ nhân tạo 11 2.1.1.1 Khái niệm .11 2.1.1.2 Phân loại .12 2.1.1.3 Ứng dụng 13 2.1.2 Chatbots .16 2.1.2.1 Khái niệm .16 2.1.2.2 Lịch sử hình thành 16 2.1.2.3 Cấu tạo chatbots 21 2.1.2.4 Phân loại .22 2.1.2.5 Cách thức hoạt động 23 2.1.2.6 Một số tảng hỗ trợ cho việc phát triển chatbots 23 2.1.2.7 Một số ứng dụng chatbot 24 2.1.2.8 Xu hướng phát triển 24 2.1.3 Xử lý ngôn ngữ tự nhiên 25 2.1.3.1 Khái niệm .25 2.1.3.2 Các bước xử lý .25 2.1.3.3 Các toán ứng dụng 25 2.1.3.4 Các toán xử lý .27 2.2 Phương pháp nghiên cứu .27 2.2.1 Phương pháp nghiên cứu lý thuyết 27 2.2.2 Phương pháp nghiên cứu thực nghiệm 28 2.2.3 Phương pháp điều tra 28 2.3 Giới thiệu giải Conditional Random Field 28 2.3.1 Giới thiệu chung 28 2.3.2 Mô tả 29 2.3.3 Biến thể 29 2.4 Giới thiệu ngôn ngữ Python .30 vii 2.4.1 Khái niệm 30 2.4.2 Lịch sử .30 2.4.3 Tính 31 2.4.4 Được ứng dụng 31 CHƯƠNG GIẢI THUẬT CRF VÀ CÁC CHIẾN LƯỢC ÁP DỤNG .33 3.1 Trích xuất tính văn tài liệu 33 3.1.1 Mã hóa .33 3.1.2 Ngưng từ 33 3.1.3 Viết hoa 33 3.1.4 Tiếng lóng từ viết tắc 33 3.1.5 Loại bỏ tiếng ồn 34 3.1.6 Sửa tả 34 3.1.7 Gốc 34 3.1.8 Bổ sung 34 3.1.9 Word2Vec 35 3.1.10 Vectơ toàn cầu cho biểu diễn từ (GloVe) .35 3.1.11 Trình bày từ ngữ theo ngữ cảnh 35 3.1.12 FastText 37 3.1.13 Tần suất kỳ hạn .37 3.1.14 Tần suất tài liệu nghịch đảo thuật ngữ 37 3.2 Giảm kích thước 38 3.3 Kỹ thuật phân loại văn 38 3.3.1 Thúc đẩy 39 3.3.2 Đóng bao 39 3.3.3 Mạng thần kinh sâu 41 3.3.4 Mạng thần kinh tái diễn (RNN) 41 3.3.5 Mạng thần kinh hợp pháp (CNN) 42 3.3.6 Mạng thần kinh chuyển đổi lặp lại (RCNN) 42 3.3.7 Học sâu đa mơ hình ngẫu nhiên (RMDL) 42 viii 3.3.8 Học sâu phân cấp cho văn (HDLTex) .43 3.4 Đánh giá 43 3.5 Văn liệu tài liệu 44 3.6 Ứng dụng phân loại văn 45 3.7 Hỗ trợ phân loại văn 46 CHƯƠNG MƠ HÌNH HỆ THỐNG .48 4.1 Tổng quan 48 4.1.1 Khái quát 48 4.1.2 Sơ lược lịch sử phát triển 48 4.1.3 Kiến trúc chung hệ thống hỏi đáp .48 4.1.3.1 Giao diện người dùng (User Interface) 49 4.1.3.2 Phân tích câu hỏi (Question Analyzer) 49 4.1.3.3 Tìm kiếm liệu (Data Retrieval) .49 4.1.3.4 Rút trích câu trả lời (Answer Extraction) 49 4.1.3.5 Chiến lược xếp hạng (Ranking) 50 4.1.3.6 Xác minh câu trả lời (Answer Verification) .50 4.1.4 Một số vấn đề quan tâm thiết kế hệ thống hỏi đáp 50 4.1.5 Các yêu cầu hệ thống hỏi đáp 50 4.2 Tiền xử lý phân loại câu hỏi .50 4.2.1 Tiền xử lý liệu .50 4.2.1.1 Phân tích câu hỏi 50 4.2.1.2 Tách từ 51 4.2.1.3 Biểu diễn văn 52 4.2.2 Phân loại câu hỏi 52 4.2.2.1 Rút trích đặc trưng văn 52 4.2.2.2 Phân loại câu hỏi tiếng Việt 54 4.3 Xây dựng mơ hình hệ thống 55 4.3.1 Lựa chọn mơ hình thuật tốn 55 4.3.2 Mơ hình kiến trúc hệ thống 56 ix 4.3.3 Cài đặt 58 CHƯƠNG KẾT QUẢ NGHIÊN CỨU 59 5.1 Hệ thống 59 5.1.1 Thao tác với liệu chuyên ngành Công Nghệ Thông Tin 59 5.1.2 Giao diện hệ thống chương trình .60 5.2 Hướng dẫn cài đặt 62 5.3 Hướng dẫn sử dụng 63 5.3.1 Thêm liệu 63 5.3.2 Thực thi chương trình 64 CHƯƠNG KẾT LUẬN 67 TÀI LIỆU THAM KHẢO .68 x DANH SÁCH BẢNG Bảng 1 Ngành nghề kinh doanh .2 Bảng Bảng sản phẩm phần mềm Bảng Bảng chi tiết dịch vụ xi DANH SÁCH HÌNH Hình Trí thơng minh nhân tạo 11 Hình 2 Mơ hình xe khơng người lái .12 Hình Cơng nghệ nhận thức 13 Hình Các hình thức giao hàng Uber 14 Hình Máy bay thiết bị không người láy 14 Hình Giải pháp giáo dục 15 Hình Cách thức hoạt động Chatbots 16 Hình Phép thử Turing 17 Hình Tương tác người dùng với ELIZA 18 Hình 10 Mơ hình cách thức hoạt động Chatbot 23 Hình Hệ thống tìm kiếm thơng tin IR 48 Hình Kiến trúc chung hệ thống hỏi đáp 49 Hình Mơ hình hoạt động hệ thống 56 Hình 4 Giai đoạn phân tích câu hỏi 56 Hình Tiến trình phân lớp 57 Hình Tiến trình tìm kiếm câu hỏi 57 Hình Tiến trình tìm kiếm câu trả lời 57 Hình Tiến trình tìm kiếm câu hỏi đến chuyên gia 57 Hình Giao diện chương trình 60 Hình Giao diện dành cho quản trị viên 60 Hình Giao diện thêm liệu .61 Hình Giao diện chọn tệp liệu cần thêm vào hệ thống 61 Hình 5 Giao diện chọn thư mục chứa tệp cần thêm vào hệ thống 62 Hình Tệp TraLoiTuDOngCNTT.exe 62 Hình Cửa sổ cài đặt 62 Hình Biểu tượng chạy chương trình Desktop 63 Hình Tệp chạy chương trình thư mục cài đặt 63 Hình 10 Menu “Thêm liệu” 63 Hình 11 Giao diện chọn đường dẫn tệp liệu cần thêm vào 64 Hình 12 Giao diện form người dùng 65 Hình 13 Giao diện sau thực thi .65 Hình 14 Giao diện nhân viên người dùng gửi yêu cầu 66 Hình 15 Giao diện người dùng nhân viên phản hồi .66 xii Tên viết tắt DANH MỤC TỪ VIẾT TẮT TNHH CSDL Tên đầy đủ NLP Trách nhiệm hữu hạn HOG Cơ sở liệu OCR Natural Language Processing CRFs Histogram of Oriented Gradients AI Optical Character Recognition Conditional Random Field Artificial intelligence xiii

Ngày đăng: 28/02/2024, 07:40

Tài liệu cùng người dùng

  • Đang cập nhật ...

Tài liệu liên quan