最近常听同事提及相关性分析,正巧看到这个google的开源库，并把相关操作与调试结果记录下来。

小说集可以百度《金庸小说全集 14部》全（TXT）作者：金庸 下载下来。
需要整理好格式，门派和武功名称之间需要有换行符，留意删除掉最后一行的空白区域。
下载完成后可以用自己习惯的工具或程序做相应调整，因语料内容太长，博客里面不允许"堆砌",所以没复制上来，有需要的可以再联系。

with open('names.txt') as f:

    data = [line.strip()  for line in f.readlines()]

novels = data[::2]

names = data[1::2]

novel_names = {k: v.split() for k, v in zip(novels, names)}

//可以在这里打印下看是不是都读取过

//开始分词并加载

for _, names in novel_names.items():#.iteritems():（Python2的旧写法）

    for name in names:

        jieba.add_word(name)

with open("kongfu.txt",encoding='UTF-8') as f:

    kungfu_names = [line.strip()

                    for line in f.readlines()]

with open("bangs.txt") as f:

    bang_names = [line.strip()

                  for line in f.readlines()]

for name in kungfu_names:

    jieba.add_word(name)

for name in bang_names:

    jieba.add_word(name)

novels = ["书剑恩仇录", "天龙八部","碧血剑","越女剑","飞狐外传","侠客行","射雕英雄传","神雕侠侣","连城诀","鸳鸯刀","倚天屠龙记","白马啸西风","笑傲江湖","雪山飞狐","鹿鼎记"]

//你知道的：飞雪连天射白鹿，笑书神侠倚碧鸳

sentences = []

for novel in novels:

    with open('{}.txt'.format(novel),encoding='UTF-8') as f:

        data = [line.strip()

                for line in f.readlines()

                if line.strip()]

    for line in data:

        words = list(jieba.cut(line))

        sentences.append(words)

model = gensim.models.Word2Vec(sentences,

                               size=200, #

                               window=5,

                               min_count=5,

                               workers=4)

//便于使用的方法
def get_gongfu(a, b, c):

    d, _ = model1.most_similar(positive=[c, b], negative=[a])[0]

    print (c,d)

//使用举例一
print ('-------------若黄蓉使出打狗棒法，郭靖会怎样？-------------------------')

get_gongfu("黄蓉","打狗棒法","郭靖")

print ('-------------若黄蓉拿着打狗棒，郭靖又会拿啥？-------------------------')

get_gongfu("黄蓉","打狗棒","郭靖")

输出结果：
-------------相关性:乔峰-------------------------

%% 虚竹 0.8226621747016907

%% 慕容复 0.809000551700592

%% 段正淳 0.808856725692749

%% 木婉清 0.7898266315460205

%% 童姥 0.7881260514259338

%% 袁承志 0.7863771915435791

%% 全冠清 0.7761101722717285

%% 谢烟客 0.7738543748855591

%% 俞莲舟 0.7663788199424744

%% 陆菲青 0.7651679515838623

-------------相关性:阿朱-------------------------

阿紫 0.8502078056335449

王语嫣 0.8323276042938232

木婉清 0.8188427090644836

方怡 0.81195068359375

钟灵 0.8042664527893066

仪琳 0.7905520796775818

青青 0.7837553024291992

香香公主 0.7774882316589355

盈盈 0.7765697836875916

马夫人 0.7628135681152344

-------------相关性:降龙十八掌-------------------------

打狗棒法 0.9099119901657104

太极拳 0.8792168498039246

空明拳 0.8742830157279968

绝招 0.864672064781189

一阳指 0.8576483726501465

蛤蟆功 0.8443030714988708

心法 0.8419612646102905

棒法 0.840523362159729

罗汉拳 0.838168740272522

小擒拿手 0.8356980085372925

-------------若黄蓉使出打狗棒法，郭靖会怎样？-------------------------

郭靖 降龙十八掌

-------------若黄蓉拿着打狗棒，郭靖又会拿啥？-------------------------

郭靖 令旗