histcite和citespace哪个引文分析能力更强？

import os
import re
import pandas as pd
def extract_all_text(folder):
    读取wos文件，抽取引文数据
    :param folder: 文件所在文件夹
    :return:
    record_list = []
    di_year_list = []
    file_list = os.listdir(folder)
    for file in file_list:
        tmp_record_list, tmp_di_year_list = extract_text(os.path.join(folder, file))
        print('{}读取完成，record数据量:{} , di数据量:{}'.format(file, len(tmp_record_list), len(tmp_di_year_list)))
        record_list.extend(tmp_record_list)
        di_year_list.extend(tmp_di_year_list)
    print('总数据量:{}'.format(len(record_list)))
    df1 = pd.DataFrame(record_list)[['DI', 'CR']]
    df1.to_excel('wos_citation.xlsx', index=False)
    df2 = pd.DataFrame(di_year_list)[['DI', 'PY']]
    df2.drop_duplicates(subset=['DI'], keep='first', inplace=True)
    df2.to_excel('di_py.xlsx', index=False)
    return df1     
def extract_text(file_name):
    record_result = []
    DI_year_result = []
    with open(file_name, 'r', encoding='utf8') as text_file:
        record = ''
        while True:
            line = text_file.readline()
            if line.startswith("PT"):
                record = ''
            if line.startswith("ER"):
                dict_result = {}
                record = record + line
                for col_name in ['CR', 'DI', 'PY']:
                    reg_str = r'\n' + col_name + '\s{1}((.*\n)+?)[0-9A-Z]{2}\s{1}'
                    p_reg = re.compile(reg_str)
                    m_reg = re.search(p_reg, record)
                    if m_reg:
                        target = m_reg.group(1)
                        if col_name in ['CR']:
                            target_origin = re.sub(r'\n\s{2,}', '#$$#', target)
                            target = ','.join([x.split(' ')[-1] for x in target_origin.split('#$$#') if 'DOI' in x])
                            DI_year_result.extend([{'DI': x.split(' ')[-1].replace('\n', ''), 'PY':x.split(',')[1].replace(' ', '')} for x in target_origin.split('#$$#') if 'DOI' in x])
                        else:
                            target = re.sub(r'\n\s{2,}', ' ', target)
                        target = re.sub(r'\n', '', target)
                        dict_result[col_name] = target
                    else:
                        dict_result[col_name] = ''
                DI_year_result.append({'DI': dict_result['DI'], 'PY': dict_result['PY']})
                record_result.append(dict_result)
            record = record + line
            if not line:
                break
    return record_result, DI_year_result
# 传入您的文件夹
record_df = extract_all_text(FOLDER_PATH_OF_DATA)

def generate_citation_network(records_df):
    生成邻接矩阵
    :param records_df: 
    :return:
    # 筛选不为空的数据
    records_df = records_df[(records_df['DI'] != '') & (records_df['CR'] != '')]
    result_list = []
    for index, item in records_df.iterrows():
        CR_list = item['CR'].split(',')
        for CR in CR_list:
            result_list.append({'s': CR, 'd': item['DI']})
    matrix_df = pd.DataFrame(result_list)
    matrix_df.to_excel('citation_network.xlsx', index=False)
    return matrix_df
matrix_df = generate_citation_network(record_df)

def draw_graph(network_df, di_ti_year_dict):
    画出引文网络图
    :param network_df: 引文网络
    :param di_ti_year_dict: DOI及其年份数据
    :return: 
    # 创建有向图
    G = nx.DiGraph()
    # 加入边的集合
    G.add_edges_from(network_df.to_records(index=False).tolist())
    # 我们通过节点度数来筛选，出度大于45，即至少被45篇文章引用
    outdeg = list(G.out_degree)
    to_keep_out = [n[0] for n in outdeg if n[1] > 45]
    print('keep out: {}'.format(len(to_keep_out)))
    G_sub_1 = G.subgraph(to_keep_out)
    # 再找到最大的联通子图（此处用若联通分量：同一弱连通分量里的任意两个点，保证至少一方能到达另一方）
    components_list = list(nx.weakly_connected_components(G_sub_1))
    count_list = [len(x) for x in components_list]
    max_count = max(count_list)
    print('最大弱联通分量的节点个数:{}'.format(max_count))
    max_index = count_list.index(max_count)
    max_components = components_list[max_index]
    sub_set = set(to_keep_out).intersection(set(max_components))
    print('最终节点数量: {}'.format(len(sub_set)))
    G_sub_2 = G.subgraph(sub_set)
    # 计算出度用于最终显示节点大小
    out_degree_dict = dict(G_sub_2.out_degree)
    year_di_group = {}
    #得到年的列表
    year_list = sorted(set([di_ti_year_dict[x] for x in sub_set]))
    print(year_list)
    # 得到每年的数据
    # 年份太多，对年份分组， 1990以前， 1991-1995， 1996-2000， 2001-2005， 2006-2010， 2011-2015， 2016-2020
    year_di_group[1990] = [x for x in sub_set if di_ti_year_dict[x] <= 1990]
    year_di_group[1995] = [x for x in sub_set if di_ti_year_dict[x] > 1990 and di_ti_year_dict[x] <= 1995]
    year_di_group[2000] = [x for x in sub_set if di_ti_year_dict[x] > 1995 and di_ti_year_dict[x] <= 2000]
    year_di_group[2005] = [x for x in sub_set if di_ti_year_dict[x] > 2000 and di_ti_year_dict[x] <= 2005]
    year_di_group[2010] = [x for x in sub_set if di_ti_year_dict[x]> 2005 and di_ti_year_dict[x] <= 2010]
    year_di_group[2015] = [x for x in sub_set if di_ti_year_dict[x]> 2010 and di_ti_year_dict[x] <= 2015]
    year_di_group[2020] = [x for x in sub_set if di_ti_year_dict[x]> 2015 and di_ti_year_dict[x] <= 2020]
    # 根据年份确定点的位置
    pos = {}
    for year, node_list in year_di_group.items():
        for index, node in enumerate(node_list):
            pos[node] = (year, index)
    #根据被引年份，修改边的颜色
    year_color = {1990: '#CCFF00', 1995: '#E9967A', 2000: '#ADD8E6', 2005: '#D3D3D3', 2010: '#66FFE6',
                  2015: '#66FF00', 2020: '#CCCCFF'}
    edge_colors = [year_color[pos[e[1]][0]] for e in G_sub_2.edges]
    # 节点标签， 显示年+doi
    label_dict = {}
    for n in sub_set:
        label_dict[n] = str(di_ti_year_dict[n]) + ' '+n
    fig, ax = plt.subplots()
    plt.axis("on")
    nx.draw_networkx(G_sub_2, pos=pos, ax=ax, labels=label_dict, font_size=6, linewidths=0.5, node_color='#FFEF33',connectionstyle="arc3,rad=0.1" ,  edge_color=edge_colors, node_size=[v * 100 for v in out_degree_dict.values()])
    ax.tick_params(left=False, bottom=True, labelleft=False, labelbottom=True)
    plt.show()
network_df = pd.read_excel('citation_network.xlsx')
di_py_df = pd.read_excel('di_py.xlsx', dtype={'PY': np.int})