Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -467,116 +467,142 @@ def self_crawl_search(query, max_results=3):
|
|
| 467 |
all_results.extend(duckduckgo_search(f"{query} 논문 학술", 2))
|
| 468 |
return all_results
|
| 469 |
def run_plagiarism(text, progress=gr.Progress()):
|
|
|
|
|
|
|
| 470 |
if not text or len(text.strip())<50:
|
| 471 |
return "<div style='padding:20px;text-align:center;color:#888;'>⚠️ 최소 50자 이상</div>", ""
|
|
|
|
| 472 |
text = text.strip()
|
| 473 |
sents = split_sentences(text)
|
| 474 |
now = datetime.now().strftime("%Y-%m-%d %H:%M")
|
| 475 |
-
|
| 476 |
-
has_gemini = bool(HAS_GENAI and GEMINI_KEY)
|
| 477 |
progress(0.05, "문장 분리...")
|
| 478 |
-
|
| 479 |
-
for i in range(0, len(sents), 4):
|
| 480 |
-
block = ' '.join(sents[i:i+4])
|
| 481 |
-
if len(block) > 20:
|
| 482 |
-
blocks.append({"text": block, "sent_indices": list(range(i, min(i+4, len(sents))))})
|
| 483 |
all_sources = []
|
| 484 |
-
sent_matches = {i: [] for i in range(len(sents))} # 문장별 매칭 정보
|
| 485 |
-
block_results = []
|
| 486 |
log_lines = []
|
| 487 |
-
if has_brave:
|
| 488 |
-
progress(0.15, f"Brave Search 병렬 검색 ({len(blocks)}블록)...")
|
| 489 |
-
queries = []
|
| 490 |
-
for b in blocks:
|
| 491 |
-
key_phrase = b["text"][:60].strip()
|
| 492 |
-
queries.append(f'"{key_phrase}"')
|
| 493 |
-
brave_results = parallel_brave_search(queries[:20])
|
| 494 |
-
for q, results in brave_results.items():
|
| 495 |
-
for r in results:
|
| 496 |
-
all_sources.append(r)
|
| 497 |
-
for b in blocks:
|
| 498 |
-
if q.strip('"') in b["text"][:60]:
|
| 499 |
-
for si in b["sent_indices"]:
|
| 500 |
-
sent_matches[si].append({"source": r["title"], "url": r["url"], "type": "Brave"})
|
| 501 |
-
log_lines.append(f"Brave Search: {len(queries)}쿼리 → {sum(len(v) for v in brave_results.values())}건")
|
| 502 |
-
else:
|
| 503 |
-
progress(0.15, f"자체 웹 검색 ({len(blocks)}블록)...")
|
| 504 |
-
crawl_queries = []
|
| 505 |
-
for b in blocks[:10]: # 최대 10블록
|
| 506 |
-
key_phrase = b["text"][:50].strip()
|
| 507 |
-
crawl_queries.append((key_phrase, b))
|
| 508 |
-
with ThreadPoolExecutor(max_workers=5) as executor:
|
| 509 |
-
futures = {executor.submit(self_crawl_search, q, 3): (q, b) for q, b in crawl_queries}
|
| 510 |
-
for future in as_completed(futures):
|
| 511 |
-
q, b = futures[future]
|
| 512 |
-
try:
|
| 513 |
-
results = future.result()
|
| 514 |
-
for r in results:
|
| 515 |
-
all_sources.append(r)
|
| 516 |
-
for si in b["sent_indices"]:
|
| 517 |
-
sent_matches[si].append({"source": r["title"], "url": r["url"], "type": r.get("source","Web")})
|
| 518 |
-
except: pass
|
| 519 |
-
log_lines.append(f"자체 웹검색: {len(crawl_queries)}쿼리 (DuckDuckGo)")
|
| 520 |
-
progress(0.40, "학술 DB 검색 (KCI/RISS/arXiv)...")
|
| 521 |
-
words = split_words(text)
|
| 522 |
-
wf = Counter(words)
|
| 523 |
-
keywords = [w for w, c in wf.most_common(20) if len(w) >= 2 and c >= 2][:5]
|
| 524 |
-
kw_query = ' '.join(keywords[:3])
|
| 525 |
-
academic_results = []
|
| 526 |
-
with ThreadPoolExecutor(max_workers=3) as executor:
|
| 527 |
-
futures = [
|
| 528 |
-
executor.submit(search_kci, kw_query),
|
| 529 |
-
executor.submit(search_riss, kw_query),
|
| 530 |
-
executor.submit(search_arxiv, kw_query),
|
| 531 |
-
]
|
| 532 |
-
for future in as_completed(futures):
|
| 533 |
-
try:
|
| 534 |
-
results = future.result()
|
| 535 |
-
academic_results.extend(results)
|
| 536 |
-
all_sources.extend(results)
|
| 537 |
-
except: pass
|
| 538 |
-
log_lines.append(f"학술DB: KCI/RISS/arXiv → {len(academic_results)}건")
|
| 539 |
-
gemini_results = []
|
| 540 |
-
if has_gemini:
|
| 541 |
-
progress(0.60, "Gemini + Google Search (전체 텍스트 검사)...")
|
| 542 |
-
# ✅ 개선: 블록 제한 제거! 전체 텍스트를 한 번에 Gemini로 검사
|
| 543 |
-
gr_result = gemini_plagiarism_check(text)
|
| 544 |
-
if gr_result:
|
| 545 |
-
gemini_results.append(gr_result)
|
| 546 |
-
for src in gr_result.get("sources", []):
|
| 547 |
-
all_sources.append(src)
|
| 548 |
-
pct = gr_result.get('pct', 0) if gr_result else 0
|
| 549 |
-
log_lines.append(f"✓ Gemini: 전체 텍스트 검사 완료 → {pct}% 유사도, {len(gr_result.get('sources',[]))}출처 발견")
|
| 550 |
-
progress(0.80, "보고서 생성...")
|
| 551 |
-
matched_sents = sum(1 for si, matches in sent_matches.items() if matches)
|
| 552 |
-
total_sents = len(sents)
|
| 553 |
|
| 554 |
-
#
|
| 555 |
-
#
|
|
|
|
|
|
|
|
|
|
| 556 |
|
| 557 |
-
if
|
| 558 |
-
|
| 559 |
-
gemini_pct = gemini_results[0]["pct"]
|
| 560 |
|
| 561 |
-
|
| 562 |
-
|
| 563 |
-
|
| 564 |
-
|
| 565 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 566 |
|
| 567 |
-
|
| 568 |
-
|
| 569 |
-
|
| 570 |
-
|
| 571 |
-
|
| 572 |
-
|
| 573 |
-
|
| 574 |
-
|
| 575 |
-
|
| 576 |
-
|
| 577 |
-
|
| 578 |
-
|
| 579 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 580 |
seen_urls = set()
|
| 581 |
unique_sources = []
|
| 582 |
for s in all_sources:
|
|
@@ -584,254 +610,79 @@ def run_plagiarism(text, progress=gr.Progress()):
|
|
| 584 |
if url and url not in seen_urls:
|
| 585 |
seen_urls.add(url)
|
| 586 |
unique_sources.append(s)
|
| 587 |
-
|
| 588 |
-
|
| 589 |
-
elif plag_pct >=
|
| 590 |
-
elif plag_pct >=
|
| 591 |
-
|
| 592 |
-
|
| 593 |
-
|
| 594 |
-
matches = sent_matches.get(i, [])
|
| 595 |
-
if matches:
|
| 596 |
-
best = matches[0]
|
| 597 |
-
sent_analysis.append({"idx":i, "text":s, "matched":True, "source":best.get("source","")[:40], "url":best.get("url",""), "type":best.get("type","")})
|
| 598 |
-
else:
|
| 599 |
-
sent_analysis.append({"idx":i, "text":s, "matched":False})
|
| 600 |
-
sim_sents = [s for s in sent_analysis if s["matched"]]
|
| 601 |
-
src_groups = {}
|
| 602 |
-
for src in unique_sources:
|
| 603 |
-
key = src.get("url","")[:80]
|
| 604 |
-
if key not in src_groups:
|
| 605 |
-
src_groups[key] = {"title":src.get("title",""), "url":src.get("url",""), "source":src.get("source",""), "count":0}
|
| 606 |
-
src_groups[key]["count"] += 1
|
| 607 |
-
src_list = sorted(src_groups.values(), key=lambda x: -x["count"])
|
| 608 |
-
methods_used = []
|
| 609 |
-
if has_brave: methods_used.append("Brave Search(병렬)")
|
| 610 |
-
elif all_sources: methods_used.append("DuckDuckGo(자체크롤링)")
|
| 611 |
-
methods_used.append("KCI · RISS · arXiv")
|
| 612 |
-
if has_gemini: methods_used.append("Gemini+Google Search")
|
| 613 |
-
method_str = " + ".join(methods_used)
|
| 614 |
-
gc = grade_color
|
| 615 |
word_count = len(split_words(text))
|
| 616 |
char_count = len(text)
|
| 617 |
doc_id = hashlib.md5(text[:100].encode()).hexdigest()[:8].upper()
|
| 618 |
-
|
| 619 |
-
citation_pct = 0
|
| 620 |
-
cat_suspect = len(sim_sents) # 의심
|
| 621 |
-
cat_cited = 0 # 인용 (형식적 인용 감지)
|
| 622 |
-
cat_normal = total_sents - cat_suspect - cat_cited # 일반
|
| 623 |
-
cat_suspect_pct = int(cat_suspect / max(1, total_sents) * 100)
|
| 624 |
-
cat_normal_pct = 100 - cat_suspect_pct
|
| 625 |
-
def src_icon(s):
|
| 626 |
-
src = s.get("source","").lower()
|
| 627 |
-
if "kci" in src: return "📚", "KCI"
|
| 628 |
-
if "riss" in src: return "📖", "RISS"
|
| 629 |
-
if "arxiv" in src: return "📄", "arXiv"
|
| 630 |
-
if "google" in src: return "🔍", "Google"
|
| 631 |
-
if "brave" in src: return "🌐", "Brave"
|
| 632 |
-
return "🌐", "Web"
|
| 633 |
src_rows = ""
|
| 634 |
-
for i,
|
| 635 |
-
|
| 636 |
-
ico, stype = src_icon(sg)
|
| 637 |
-
title_short = sg["title"][:50] or "(제목 없음)"
|
| 638 |
-
url_short = sg["url"][:60]
|
| 639 |
src_rows += f"""<tr>
|
| 640 |
-
<td style="padding:6px 8px;font-size:11px;text-align:center;border:1px solid #D5D5D5;
|
| 641 |
-
<td style="padding:6px 8px;font-size:11px;text-align:center;border:1px solid #D5D5D5;">
|
| 642 |
-
<td style="padding:6px 8px;border:1px solid #D5D5D5;
|
| 643 |
-
<td style="padding:6px 8px;
|
| 644 |
-
<td style="padding:6px 10px;border:1px solid #D5D5D5;"><div style="background:#EDEDED;height:14px;border-radius:2px;overflow:hidden;"><div style="background:linear-gradient(90deg,#D63031,#FF7675);height:100%;width:{max(3,pct)}%;border-radius:2px;"></div></div></td>
|
| 645 |
-
</tr>"""
|
| 646 |
-
suspect_rows = ""
|
| 647 |
-
for i, sa in enumerate(sim_sents[:15]):
|
| 648 |
-
suspect_rows += f"""<tr>
|
| 649 |
-
<td style="padding:8px;font-size:11px;text-align:center;color:#888;border:1px solid #D5D5D5;vertical-align:top;">{i+1}</td>
|
| 650 |
-
<td style="padding:8px;font-size:11px;line-height:1.7;border:1px solid #D5D5D5;vertical-align:top;"><span style="background:#FFF3CD;border-bottom:2px solid #FFD43B;padding:1px 3px;">{sa["text"][:90]}</span></td>
|
| 651 |
-
<td style="padding:8px;font-size:10px;line-height:1.6;border:1px solid #D5D5D5;vertical-align:top;color:#555;"><span style="background:#FFE0E0;border-bottom:2px solid #E74C3C;padding:1px 3px;">{sa["text"][:70]}...</span></td>
|
| 652 |
-
<td style="padding:8px;font-size:10px;border:1px solid #D5D5D5;vertical-align:top;"><a href="{sa.get('url','#')}" target="_blank" style="color:#2E86C1;text-decoration:none;font-weight:600;">{sa["source"][:28]}</a><br><span style="font-size:8px;color:#AAA;">{sa.get('type','')}</span></td>
|
| 653 |
</tr>"""
|
| 654 |
-
|
| 655 |
-
|
| 656 |
-
|
| 657 |
-
if sa["matched"]:
|
| 658 |
-
full_hl += f'<span style="background:#FFD6D6;border-bottom:2px solid #E74C3C;padding:1px 2px;cursor:pointer;" title="[의심 #{sidx}] 출처: {sa.get("source","")}">{sa["text"]}</span> '
|
| 659 |
-
else:
|
| 660 |
-
full_hl += f'<span style="color:#333;">{sa["text"]}</span> '
|
| 661 |
-
bar_suspect_w = max(2, cat_suspect_pct) if cat_suspect > 0 else 0
|
| 662 |
-
bar_normal_w = 100 - bar_suspect_w
|
| 663 |
-
HDR_BG = '#3B7DD8'
|
| 664 |
-
HDR_BG2 = '#4A8DE0'
|
| 665 |
-
TH = 'padding:8px 10px;font-size:10px;font-weight:700;color:#fff;background:{};text-align:center;border:1px solid {};'.format(HDR_BG, HDR_BG)
|
| 666 |
-
TL = 'padding:7px 10px;font-size:11px;color:#444;font-weight:600;background:#EDF2FA;border:1px solid #D5D5D5;'
|
| 667 |
TV = 'padding:7px 10px;font-size:12px;color:#333;border:1px solid #D5D5D5;'
|
| 668 |
-
|
| 669 |
-
html = f"""<div style="font-family:'Noto Sans KR',
|
| 670 |
-
<
|
| 671 |
-
|
| 672 |
-
<
|
| 673 |
-
<td>
|
| 674 |
-
<div style="font-size:10px;opacity:0.8;letter-spacing:1px;margin-bottom:4px;">AI TEXT DETECTOR · PLAGIARISM REPORT</div>
|
| 675 |
-
<div style="font-size:22px;font-weight:900;letter-spacing:-0.5px;">표절 검사 결과 확인서</div>
|
| 676 |
-
</td>
|
| 677 |
-
<td style="text-align:right;vertical-align:bottom;">
|
| 678 |
-
<div style="font-size:10px;opacity:0.7;">문서번호 {doc_id}</div>
|
| 679 |
-
<div style="font-size:10px;opacity:0.7;">{now}</div>
|
| 680 |
-
</td>
|
| 681 |
-
</tr></table>
|
| 682 |
-
</div>
|
| 683 |
-
<!-- ═══════ 제출 정보 테이블 ═══════ -->
|
| 684 |
-
<div style="padding:16px 24px 0;">
|
| 685 |
-
<div style="{SEC}">📋 검사 정보</div>
|
| 686 |
-
<table style="width:100%;border-collapse:collapse;">
|
| 687 |
-
<tr>
|
| 688 |
-
<td style="{TL}width:90px;">검사 일시</td>
|
| 689 |
-
<td style="{TV}">{now}</td>
|
| 690 |
-
<td style="{TL}width:90px;">문서번호</td>
|
| 691 |
-
<td style="{TV}">{doc_id}</td>
|
| 692 |
-
</tr>
|
| 693 |
-
<tr>
|
| 694 |
-
<td style="{TL}">검사 방법</td>
|
| 695 |
-
<td style="{TV}font-size:10px;" colspan="3">{method_str}</td>
|
| 696 |
-
</tr>
|
| 697 |
-
<tr>
|
| 698 |
-
<td style="{TL}">전체 분량</td>
|
| 699 |
-
<td style="{TV}" colspan="3">글자수 <b>{char_count:,}</b> · 어절수 <b>{word_count:,}</b> · 문장수 <b>{total_sents}</b></td>
|
| 700 |
-
</tr>
|
| 701 |
-
<tr>
|
| 702 |
-
<td style="{TL}">검색 범위</td>
|
| 703 |
-
<td style="{TV}" colspan="3">인터넷(웹), 학술논문(KCI·RISS), 해외논문(arXiv), Google Scholar</td>
|
| 704 |
-
</tr>
|
| 705 |
-
</table>
|
| 706 |
</div>
|
| 707 |
-
<!-- ═══════ 검사 결과 (도넛 + 바) ═══════ -->
|
| 708 |
<div style="padding:18px 24px 0;">
|
| 709 |
-
<div style="{SEC}">📊 검사 결과</div>
|
| 710 |
<table style="width:100%;border-collapse:collapse;">
|
| 711 |
<tr>
|
| 712 |
-
<
|
| 713 |
-
|
| 714 |
-
<div style="
|
| 715 |
-
<svg viewBox="0 0 36 36" style="width:140px;height:140px;transform:rotate(-90deg);">
|
| 716 |
-
<circle cx="18" cy="18" r="15.9" fill="none" stroke="#E8ECF0" stroke-width="2.8"/>
|
| 717 |
-
<circle cx="18" cy="18" r="15.9" fill="none" stroke="{gc}" stroke-width="2.8" stroke-dasharray="{plag_pct} {100-plag_pct}" stroke-linecap="round"/>
|
| 718 |
-
</svg>
|
| 719 |
-
<div style="position:absolute;top:50%;left:50%;transform:translate(-50%,-50%);text-align:center;">
|
| 720 |
-
<div style="font-size:9px;color:#888;margin-bottom:2px;">표절률</div>
|
| 721 |
-
<div style="font-size:32px;font-weight:900;color:{gc};line-height:1;">{plag_pct}<span style="font-size:14px;font-weight:700;">%</span></div>
|
| 722 |
-
</div>
|
| 723 |
-
</div>
|
| 724 |
-
<div style="margin-top:8px;padding:4px 12px;background:{grade_bg};border:1px solid {gc};border-radius:20px;display:inline-block;">
|
| 725 |
-
<span style="font-size:11px;font-weight:800;color:{gc};">{grade}</span>
|
| 726 |
-
</div>
|
| 727 |
</td>
|
| 728 |
-
<
|
| 729 |
-
|
| 730 |
-
<
|
| 731 |
-
|
| 732 |
-
<div
|
| 733 |
-
|
| 734 |
-
<div style="background:#E8ECF0;width:{bar_normal_w}%;"></div>
|
| 735 |
-
</div>
|
| 736 |
-
<div style="display:flex;gap:14px;flex-wrap:wrap;font-size:9px;">
|
| 737 |
-
<span style="display:flex;align-items:center;gap:3px;"><span style="display:inline-block;width:10px;height:10px;background:#E74C3C;border-radius:2px;"></span> 의심 <b>{cat_suspect}</b>건</span>
|
| 738 |
-
<span style="display:flex;align-items:center;gap:3px;"><span style="display:inline-block;width:10px;height:10px;background:#F39C12;border-radius:2px;"></span> 출처표시 <b>0</b>건</span>
|
| 739 |
-
<span style="display:flex;align-items:center;gap:3px;"><span style="display:inline-block;width:10px;height:10px;background:#3498DB;border-radius:2px;"></span> 인용 <b>{cat_cited}</b>건</span>
|
| 740 |
-
<span style="display:flex;align-items:center;gap:3px;"><span style="display:inline-block;width:10px;height:10px;background:#E8ECF0;border:1px solid #CCC;border-radius:2px;"></span> 일반 <b>{cat_normal}</b>건</span>
|
| 741 |
-
</div>
|
| 742 |
-
</div>
|
| 743 |
-
<!-- 비율 바 3줄 -->
|
| 744 |
-
<div style="padding:12px 18px;">
|
| 745 |
-
<div style="margin-bottom:10px;">
|
| 746 |
-
<div style="display:flex;justify-content:space-between;margin-bottom:4px;"><span style="font-size:11px;font-weight:700;color:#E74C3C;">■ 표절률</span><span style="font-size:18px;font-weight:900;color:#E74C3C;">{plag_pct}%</span></div>
|
| 747 |
-
<div style="background:#EDEDED;height:16px;border-radius:3px;overflow:hidden;border:1px solid #DDD;"><div style="background:linear-gradient(90deg,#E74C3C,#FF7675);height:100%;width:{max(1,plag_pct)}%;transition:width 0.5s;"></div></div>
|
| 748 |
-
</div>
|
| 749 |
-
<div style="margin-bottom:10px;">
|
| 750 |
-
<div style="display:flex;justify-content:space-between;margin-bottom:4px;"><span style="font-size:11px;font-weight:700;color:#F39C12;">■ 유사율</span><span style="font-size:18px;font-weight:900;color:#F39C12;">{similarity_pct}%</span></div>
|
| 751 |
-
<div style="background:#EDEDED;height:16px;border-radius:3px;overflow:hidden;border:1px solid #DDD;"><div style="background:linear-gradient(90deg,#F39C12,#FFEAA7);height:100%;width:{max(1,similarity_pct)}%;transition:width 0.5s;"></div></div>
|
| 752 |
-
</div>
|
| 753 |
-
<div>
|
| 754 |
-
<div style="display:flex;justify-content:space-between;margin-bottom:4px;"><span style="font-size:11px;font-weight:700;color:#3498DB;">■ 인용률</span><span style="font-size:18px;font-weight:900;color:#3498DB;">{citation_pct}%</span></div>
|
| 755 |
-
<div style="background:#EDEDED;height:16px;border-radius:3px;overflow:hidden;border:1px solid #DDD;"><div style="background:linear-gradient(90deg,#3498DB,#85C1E9);height:100%;width:{max(1,citation_pct)}%;transition:width 0.5s;"></div></div>
|
| 756 |
-
</div>
|
| 757 |
-
</div>
|
| 758 |
-
<!-- 요약 수치 -->
|
| 759 |
-
<div style="padding:8px 18px;background:#F8F9FB;border-top:1px solid #E8E8E8;">
|
| 760 |
-
<table style="width:100%;border-collapse:collapse;">
|
| 761 |
-
<tr>
|
| 762 |
-
<td style="padding:4px;font-size:10px;color:#888;">의심문장</td>
|
| 763 |
-
<td style="padding:4px;font-size:12px;font-weight:800;color:#E74C3C;">{cat_suspect}건</td>
|
| 764 |
-
<td style="padding:4px;font-size:10px;color:#888;">일반문장</td>
|
| 765 |
-
<td style="padding:4px;font-size:12px;font-weight:800;color:#27AE60;">{cat_normal}건</td>
|
| 766 |
-
<td style="padding:4px;font-size:10px;color:#888;">전체</td>
|
| 767 |
-
<td style="padding:4px;font-size:12px;font-weight:800;color:#333;">{total_sents}건</td>
|
| 768 |
-
</tr>
|
| 769 |
-
</table>
|
| 770 |
</div>
|
| 771 |
</td>
|
| 772 |
</tr>
|
| 773 |
</table>
|
| 774 |
</div>
|
| 775 |
-
<
|
| 776 |
-
|
| 777 |
-
<
|
| 778 |
-
|
| 779 |
-
|
| 780 |
-
|
| 781 |
-
|
| 782 |
-
|
| 783 |
-
|
| 784 |
-
|
| 785 |
-
|
| 786 |
-
|
| 787 |
-
|
| 788 |
-
|
| 789 |
-
<div style="{SEC}">🔗 표절 의심 출처 ({len(src_list)}건)</div>
|
| 790 |
-
<table style="width:100%;border-collapse:collapse;">
|
| 791 |
-
<tr>
|
| 792 |
-
<th style="{TH}width:32px;">No</th>
|
| 793 |
-
<th style="{TH}width:48px;">유형</th>
|
| 794 |
-
<th style="{TH}">출처명 / URL</th>
|
| 795 |
-
<th style="{TH}width:55px;">유사율</th>
|
| 796 |
-
<th style="{TH}width:100px;">분포</th>
|
| 797 |
-
</tr>
|
| 798 |
-
{src_rows if src_rows else '<tr><td colspan="5" style="padding:18px;text-align:center;color:#999;font-size:11px;border:1px solid #D5D5D5;background:#FAFBFC;">발견된 유사 출처가 없습니다.</td></tr>'}
|
| 799 |
-
</table>
|
| 800 |
-
</div>
|
| 801 |
-
<!-- ═══════ 의심 문장 비교 ═══════ -->
|
| 802 |
-
<div style="padding:18px 24px 0;">
|
| 803 |
-
<div style="{SEC}">⚠️ 의심 문장 비교 ({len(sim_sents)}건)</div>
|
| 804 |
-
<table style="width:100%;border-collapse:collapse;">
|
| 805 |
-
<tr>
|
| 806 |
-
<th style="{TH}width:32px;">No</th>
|
| 807 |
-
<th style="{TH}width:38%;">검사 문장 (원문)</th>
|
| 808 |
-
<th style="{TH}width:34%;">비교 문장 (출처)</th>
|
| 809 |
-
<th style="{TH}">출처</th>
|
| 810 |
-
</tr>
|
| 811 |
-
{suspect_rows if suspect_rows else '<tr><td colspan="4" style="padding:18px;text-align:center;color:#999;font-size:11px;border:1px solid #D5D5D5;background:#FAFBFC;">유사 의심 문장이 발견되지 않았습니다.</td></tr>'}
|
| 812 |
</table>
|
| 813 |
</div>
|
| 814 |
-
<
|
| 815 |
-
|
| 816 |
-
<
|
| 817 |
-
· 본 보고서는 <b>{method_str}</b> 기반 자동 표절 검사 결과입니다.<br>
|
| 818 |
-
· 검색 범위: 인터넷 웹페이지, 학술논문(KCI, RISS), 해외논문(arXiv)<br>
|
| 819 |
-
· 유사도는 문장 단위 매칭 기반이며, 최종 판정은 ���수자/검토자의 확인이 필요합니다.<br>
|
| 820 |
-
· 인용 표기(따옴표, 각주 등)가 포함된 문장은 인용으로 분류될 수 있습니다.
|
| 821 |
-
</div>
|
| 822 |
-
<!-- ═══════ 푸터 (CopyKiller 스타일) ═══════ -->
|
| 823 |
-
<div style="padding:10px 24px;background:#F0F3F8;border-top:2px solid {HDR_BG};display:flex;justify-content:space-between;align-items:center;">
|
| 824 |
-
<div>
|
| 825 |
-
<span style="font-size:13px;font-weight:900;color:{HDR_BG};">AI Detector</span>
|
| 826 |
-
<span style="font-size:9px;color:#999;margin-left:6px;">Plagiarism Checker v3.5</span>
|
| 827 |
-
</div>
|
| 828 |
-
<div style="text-align:right;">
|
| 829 |
-
<div style="font-size:9px;color:#AAA;">Powered by Brave · KCI · RISS · arXiv · Gemini</div>
|
| 830 |
-
<div style="font-size:8px;color:#CCC;">{now} · ID: {doc_id} · All Rights Reserved.</div>
|
| 831 |
-
</div>
|
| 832 |
</div>
|
| 833 |
</div>"""
|
| 834 |
-
|
|
|
|
|
|
|
|
|
|
| 835 |
return html, log
|
| 836 |
def run_detection(text, progress=gr.Progress()):
|
| 837 |
if not text or len(text.strip())<50: return "<div style='padding:20px;text-align:center;color:#888;'>⚠️ 최소 50자</div>",""
|
|
|
|
| 467 |
all_results.extend(duckduckgo_search(f"{query} 논문 학술", 2))
|
| 468 |
return all_results
|
| 469 |
def run_plagiarism(text, progress=gr.Progress()):
|
| 470 |
+
"""✅ Gemini Google Search 90% (메인) + Brave/arXiv 10% (보조)"""
|
| 471 |
+
|
| 472 |
if not text or len(text.strip())<50:
|
| 473 |
return "<div style='padding:20px;text-align:center;color:#888;'>⚠️ 최소 50자 이상</div>", ""
|
| 474 |
+
|
| 475 |
text = text.strip()
|
| 476 |
sents = split_sentences(text)
|
| 477 |
now = datetime.now().strftime("%Y-%m-%d %H:%M")
|
| 478 |
+
|
|
|
|
| 479 |
progress(0.05, "문장 분리...")
|
| 480 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
| 481 |
all_sources = []
|
|
|
|
|
|
|
| 482 |
log_lines = []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 483 |
|
| 484 |
+
# ============================================
|
| 485 |
+
# 메인 (90%): Gemini Google Search
|
| 486 |
+
# ============================================
|
| 487 |
+
gemini_pct = 0
|
| 488 |
+
gemini_sources = []
|
| 489 |
|
| 490 |
+
if HAS_GENAI and GEMINI_KEY:
|
| 491 |
+
progress(0.20, "Gemini + Google Search (메인 90% 표절 검사)...")
|
|
|
|
| 492 |
|
| 493 |
+
try:
|
| 494 |
+
client = genai.Client(api_key=GEMINI_KEY)
|
| 495 |
+
|
| 496 |
+
prompt = f"""당신은 고급 표절 검사 전문가입니다.
|
| 497 |
+
|
| 498 |
+
아래 텍스트를 Google Search로 철저히 검색하여 유사한 내용을 모두 찾으세요.
|
| 499 |
+
|
| 500 |
+
[검사 텍스트]
|
| 501 |
+
{text}
|
| 502 |
+
|
| 503 |
+
응답 형식:
|
| 504 |
+
발견된 유사 내용:
|
| 505 |
+
- [제목 1] (유사도: XX%)
|
| 506 |
+
- [제목 2] (유사도: XX%)
|
| 507 |
+
|
| 508 |
+
최종 표절율: XX%"""
|
| 509 |
+
|
| 510 |
+
contents = [
|
| 511 |
+
types.Content(
|
| 512 |
+
role="user",
|
| 513 |
+
parts=[
|
| 514 |
+
types.Part.from_text(text=prompt),
|
| 515 |
+
],
|
| 516 |
+
)
|
| 517 |
+
]
|
| 518 |
+
|
| 519 |
+
tools = [
|
| 520 |
+
types.Tool(googleSearch=types.GoogleSearch()),
|
| 521 |
+
]
|
| 522 |
+
|
| 523 |
+
config = types.GenerateContentConfig(
|
| 524 |
+
thinking_config=types.ThinkingConfig(thinking_budget=0),
|
| 525 |
+
tools=tools,
|
| 526 |
+
temperature=0.3,
|
| 527 |
+
max_output_tokens=3000,
|
| 528 |
+
)
|
| 529 |
+
|
| 530 |
+
# 스트리밍으로 수집
|
| 531 |
+
full_response = ""
|
| 532 |
+
for chunk in client.models.generate_content_stream(
|
| 533 |
+
model="gemini-2.0-flash-lite-latest",
|
| 534 |
+
contents=contents,
|
| 535 |
+
config=config,
|
| 536 |
+
):
|
| 537 |
+
if chunk.text:
|
| 538 |
+
full_response += chunk.text
|
| 539 |
+
|
| 540 |
+
# 유사도 추출
|
| 541 |
+
pm = re.search(r'(?:표절|유사도)[:\s]*(\d+)', full_response)
|
| 542 |
+
if pm:
|
| 543 |
+
gemini_pct = int(pm.group(1))
|
| 544 |
+
|
| 545 |
+
# 출처 추출
|
| 546 |
+
for m in re.finditer(r'https?://[^\s\)]+', full_response):
|
| 547 |
+
url = m.group(0)
|
| 548 |
+
if len(url) > 10:
|
| 549 |
+
gemini_sources.append({
|
| 550 |
+
"title": url.split('/')[2] if '/' in url else url,
|
| 551 |
+
"url": url,
|
| 552 |
+
"source": "Google",
|
| 553 |
+
})
|
| 554 |
+
all_sources.append({
|
| 555 |
+
"title": url.split('/')[2] if '/' in url else url,
|
| 556 |
+
"url": url,
|
| 557 |
+
"source": "Google",
|
| 558 |
+
})
|
| 559 |
+
|
| 560 |
+
log_lines.append(f"✅ Gemini: {gemini_pct}% → {len(gemini_sources)}개 출처")
|
| 561 |
+
|
| 562 |
+
except Exception as e:
|
| 563 |
+
log_lines.append(f"⚠️ Gemini: {str(e)[:80]}")
|
| 564 |
+
|
| 565 |
+
# ============================================
|
| 566 |
+
# 보조 (10%): Brave + arXiv
|
| 567 |
+
# ============================================
|
| 568 |
+
brave_pct = 0
|
| 569 |
+
arxiv_pct = 0
|
| 570 |
+
|
| 571 |
+
progress(0.60, "보조 검사 (Brave + arXiv)...")
|
| 572 |
+
|
| 573 |
+
try:
|
| 574 |
+
if BRAVE_KEY:
|
| 575 |
+
words = split_words(text)
|
| 576 |
+
key_query = ' '.join(words[:5])
|
| 577 |
+
brave_results = brave_search(key_query, 3)
|
| 578 |
+
if brave_results:
|
| 579 |
+
brave_pct = 30
|
| 580 |
+
all_sources.extend(brave_results)
|
| 581 |
+
log_lines.append(f"Brave: {len(brave_results)}건")
|
| 582 |
|
| 583 |
+
words = split_words(text)
|
| 584 |
+
wf = Counter(words)
|
| 585 |
+
keywords = [w for w, c in wf.most_common(10) if len(w) >= 3][:3]
|
| 586 |
+
arxiv_results = search_arxiv(' '.join(keywords))
|
| 587 |
+
if arxiv_results:
|
| 588 |
+
arxiv_pct = 20
|
| 589 |
+
all_sources.extend(arxiv_results)
|
| 590 |
+
log_lines.append(f"arXiv: {len(arxiv_results)}건")
|
| 591 |
+
except:
|
| 592 |
+
pass
|
| 593 |
+
|
| 594 |
+
# ============================================
|
| 595 |
+
# 최종: Gemini 90% + 보조 10%
|
| 596 |
+
# ============================================
|
| 597 |
+
progress(0.80, "보고서 생성...")
|
| 598 |
+
|
| 599 |
+
helper_pct = max(brave_pct, arxiv_pct)
|
| 600 |
+
plag_pct = int(gemini_pct * 0.9 + helper_pct * 0.1) if gemini_pct > 0 else helper_pct
|
| 601 |
+
|
| 602 |
+
# ============================================
|
| 603 |
+
# HTML 보고서
|
| 604 |
+
# ============================================
|
| 605 |
+
|
| 606 |
seen_urls = set()
|
| 607 |
unique_sources = []
|
| 608 |
for s in all_sources:
|
|
|
|
| 610 |
if url and url not in seen_urls:
|
| 611 |
seen_urls.add(url)
|
| 612 |
unique_sources.append(s)
|
| 613 |
+
|
| 614 |
+
if plag_pct >= 50: grade, gc = "표절 의심", "#FF4444"
|
| 615 |
+
elif plag_pct >= 30: grade, gc = "주의 필요", "#FF8800"
|
| 616 |
+
elif plag_pct >= 15: grade, gc = "유사 표현 일부", "#DDAA00"
|
| 617 |
+
elif plag_pct >= 5: grade, gc = "양호", "#4ECDC4"
|
| 618 |
+
else: grade, gc = "우수", "#22AA44"
|
| 619 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 620 |
word_count = len(split_words(text))
|
| 621 |
char_count = len(text)
|
| 622 |
doc_id = hashlib.md5(text[:100].encode()).hexdigest()[:8].upper()
|
| 623 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 624 |
src_rows = ""
|
| 625 |
+
for i, s in enumerate(unique_sources[:15]):
|
| 626 |
+
ico = "🔍" if "google" in s.get("source","").lower() else "🌐"
|
|
|
|
|
|
|
|
|
|
| 627 |
src_rows += f"""<tr>
|
| 628 |
+
<td style="padding:6px 8px;font-size:11px;text-align:center;border:1px solid #D5D5D5;">{i+1}</td>
|
| 629 |
+
<td style="padding:6px 8px;font-size:11px;text-align:center;border:1px solid #D5D5D5;">{ico}</td>
|
| 630 |
+
<td style="padding:6px 8px;border:1px solid #D5D5D5;font-size:11px;color:#1A3C6E;"><strong>{s['title'][:40]}</strong></td>
|
| 631 |
+
<td style="padding:6px 8px;border:1px solid #D5D5D5;font-size:9px;color:#666;word-break:break-all;">{s['url'][:50]}</td>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 632 |
</tr>"""
|
| 633 |
+
|
| 634 |
+
HDR = '#3B7DD8'
|
| 635 |
+
TL = f'padding:7px 10px;font-size:11px;color:#444;font-weight:600;background:#EDF2FA;border:1px solid #D5D5D5;'
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 636 |
TV = 'padding:7px 10px;font-size:12px;color:#333;border:1px solid #D5D5D5;'
|
| 637 |
+
|
| 638 |
+
html = f"""<div style="font-family:'Noto Sans KR',sans-serif;max-width:780px;margin:0 auto;background:#fff;border:2px solid {HDR};box-shadow:0 2px 12px rgba(0,0,0,0.08);">
|
| 639 |
+
<div style="background:linear-gradient(135deg,{HDR},#4A8DE0);padding:18px 24px;color:#fff;">
|
| 640 |
+
<div style="font-size:22px;font-weight:900;">표절 검사 결과</div>
|
| 641 |
+
<div style="font-size:10px;opacity:0.8;">문서: {doc_id} | {now}</div>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 642 |
</div>
|
|
|
|
| 643 |
<div style="padding:18px 24px 0;">
|
|
|
|
| 644 |
<table style="width:100%;border-collapse:collapse;">
|
| 645 |
<tr>
|
| 646 |
+
<td style="width:200px;padding:20px;text-align:center;border:1px solid #D5D5D5;background:#FAFBFE;">
|
| 647 |
+
<div style="font-size:48px;font-weight:900;color:{gc};">{plag_pct}%</div>
|
| 648 |
+
<div style="font-size:12px;color:#666;margin-top:4px;">표절율</div>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 649 |
</td>
|
| 650 |
+
<td style="padding:20px;border:1px solid #D5D5D5;">
|
| 651 |
+
<div style="font-size:16px;font-weight:900;color:{gc};margin-bottom:12px;">{grade}</div>
|
| 652 |
+
<div style="font-size:11px;color:#666;line-height:1.8;">
|
| 653 |
+
<div>글자수: <b>{char_count:,}</b></div>
|
| 654 |
+
<div>단어수: <b>{word_count:,}</b></div>
|
| 655 |
+
<div>출처: <b>{len(unique_sources)}</b>개</div>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 656 |
</div>
|
| 657 |
</td>
|
| 658 |
</tr>
|
| 659 |
</table>
|
| 660 |
</div>
|
| 661 |
+
<div style="padding:16px 24px;">
|
| 662 |
+
<div style="font-size:13px;font-weight:800;color:#1A3C6E;margin-bottom:10px;padding-bottom:8px;border-bottom:2px solid {HDR};">🔍 발견된 출처</div>
|
| 663 |
+
<table style="width:100%;border-collapse:collapse;font-size:10px;">
|
| 664 |
+
<thead>
|
| 665 |
+
<tr style="background:{HDR};color:white;">
|
| 666 |
+
<th style="padding:8px;border:1px solid #D5D5D5;width:30px;">순위</th>
|
| 667 |
+
<th style="padding:8px;border:1px solid #D5D5D5;width:30px;">출처</th>
|
| 668 |
+
<th style="padding:8px;border:1px solid #D5D5D5;">제목</th>
|
| 669 |
+
<th style="padding:8px;border:1px solid #D5D5D5;">URL</th>
|
| 670 |
+
</tr>
|
| 671 |
+
</thead>
|
| 672 |
+
<tbody>
|
| 673 |
+
{src_rows if src_rows else '<tr><td colspan="4" style="padding:16px;text-align:center;color:#999;">발견된 출처 없음</td></tr>'}
|
| 674 |
+
</tbody>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 675 |
</table>
|
| 676 |
</div>
|
| 677 |
+
<div style="padding:16px 24px;background:#FFF8E1;border-top:1px solid #D5D5D5;font-size:10px;color:#666;line-height:1.6;">
|
| 678 |
+
<strong>분석:</strong> <strong style="color:#D63031;">Gemini Google Search 90% (메인)</strong> + Brave/arXiv 10% (보조)<br>
|
| 679 |
+
<strong>방법:</strong> Gemini의 AI 분석 + Google Search Grounding으로 실시간 웹 검색
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 680 |
</div>
|
| 681 |
</div>"""
|
| 682 |
+
|
| 683 |
+
log = "\n".join(log_lines)
|
| 684 |
+
progress(0.95, "완료...")
|
| 685 |
+
|
| 686 |
return html, log
|
| 687 |
def run_detection(text, progress=gr.Progress()):
|
| 688 |
if not text or len(text.strip())<50: return "<div style='padding:20px;text-align:center;color:#888;'>⚠️ 최소 50자</div>",""
|